Views of the NIH

CLINICAL EPIDEMIOLOGY www.jasn.org

Genetic Variants Associated with Circulating Fibroblast Growth Factor 23

Cassianne Robinson-Cohen ,1 Traci M. Bartz,2 Dongbing Lai,3 T. Alp Ikizler,1 Munro Peacock,4 Erik A. Imel,4 Erin D. Michos,5 Tatiana M. Foroud,3 Kristina Akesson,6,7 Kent D. Taylor,8 Linnea Malmgren,6,7 Kunihiro Matsushita,5,9,10 Maria Nethander,11 Joel Eriksson,12 Claes Ohlsson,12 Daniel Mellström,12 Myles Wolf,13 Osten Ljunggren,14 Fiona McGuigan,6,7 Jerome I. Rotter,8 Magnus Karlsson,6,7 Michael J. Econs,3,4 Joachim H. Ix,15,16 Pamela L. Lutsey,17 Bruce M. Psaty,18,19 Ian H. de Boer ,20 and Bryan R. Kestenbaum 20

Due to the number of contributing authors, the afﬁliations are listed at the end of this article.

ABSTRACT Background Fibroblast growth factor 23 (FGF23), a bone-derived hormone that regulates phosphorus and vitamin D metabolism, contributes to the pathogenesis of mineral and bone disorders in CKD and is an emerging cardiovascular risk factor. Central elements of FGF23 regulation remain incompletely understood; genetic variation may help explain interindividual differences. Methods We performed a meta-analysis of genome-wide association studies of circulating FGF23 concentrations among 16,624 participants of European ancestry from seven cohort studies, excluding participants with eGFR,30 ml/min per 1.73 m2 to focus on FGF23 under normal conditions. We evaluated the association of single-nucleotide polymorphisms (SNPs) with natural log–transformed FGF23 concentration, adjusted for age, sex, study site, and principal components of ancestry. A second model additionally adjusted for BMI and eGFR. Results We discovered 154 SNPs from ﬁve independent regions associated with FGF23 concentration. 2 The SNP with the strongest association, rs17216707 (P=3.0310 24), lies upstream of CYP24A1,which encodes the primary catabolic enzyme for 1,25-dihydroxyvitamin D and 25-hydroxyvitamin D. Each additional copy of the T allele at this locus is associated with 5% higher FGF23 concentration. Another locus strongly associated with variations in FGF23 concentration is rs11741640, within RGS14 and upstream of SLC34A1 (a gene involved in renal phosphate transport). Additional adjustment for BMI and eGFR did not materially alter the magnitude of these associations. Another top locus (within ABO, the ABO blood group transferase gene) was no longer statistically signiﬁcant at the genome-wide level. Conclusions Common genetic variants located near genes involved in vitamin D metabolism and renal phosphate transport are associated with differences in circulating FGF23 concentrations.

J Am Soc Nephrol 29: 2583–2592, 2018. doi: https://doi.org/10.1681/ASN.2018020192

Fibroblast growth factor 23 (FGF23) is secreted Received February 20, 2018. Accepted August 6, 2018. from bone and plays a role in both phosphate homeostasis and the vitamin D endocrine system. Published online ahead of print. Publication date available at www.jasn.org. FGF23 induces urinary phosphate excretion by suppressing the expression of the sodium phos- Correspondence: Dr. Cassianne Robinson-Cohen, Division of Nephrology, Department of Medicine, Vanderbilt University phate cotransporter in kidney proximal tubular Medical Center, Nashville, TN 37232. Email: cassianne.robinson- cells. FGF23 also suppresses the synthesis of calci- [email protected] triol [1,25(OH)2D] by inhibiting its production via Copyright © 2018 by the American Society of Nephrology

J Am Soc Nephrol 29: 2583–2592, 2018 ISSN : 1046-6673/2910-2583 2583 CLINICAL EPIDEMIOLOGY www.jasn.org

1,25a-hydroxylase (CYP27B1) and stimulating its degrada- Significance Statement tion via 1,24,25-hydroxylase (CYP24A1).1,2 In turn, calcitriol directly enhances the transcription of FGF23 to complete a Fibroblast growth factor 23 (FGF23), a bone-derived hormone that feedback loop. The effects of FGF23 on phosphate and vitamin regulates phosphorus and vitamin D metabolism, contributes to D homeostasis require klotho as an obligatory coligand, which the pathogenesis of mineral and bone disorders in CKD and is an fi emerging cardiovascular risk factor. The authors performed a meta- activates broblast growth factor receptors and their down- analysis of genome-wide association studies of circulating FGF23 stream signaling molecules.3 Optimal phosphate balance is concentrations among 16,624 individuals of European ancestry important for many physiologic functions, from cell signaling from seven cohort studies. Afteradjusting for age,sex, studysite,and to energy metabolism and skeletal mineralization. In humans, principal components of ancestry, they found that common genetic high FGF23 concentrations contribute to the pathogenesis of variants are associated with differences in circulating FGF23 concentration; several are closely linked with enzymes, transporters, and mineral and bone disorders in CKD and are associated with receptors known to be critical to vitamin D metabolism and regula- ventricular hypertrophy, cardiovascular events, and prema- tion of phosphate levels. Future study of such variants may help ture death.4–12 illuminate the mechanism and clinical implications of FGF23’s role in Despite intense investigation, central elements of FGF23 reg- vitamin D and phosphate homeostasis. ulation remain unknown. The primary role of FGF23 in phosphate homeostasis suggests regulation by serum phosphate excluded participants with eGFR,30 ml/min per 1.73 m2. concentrations; however, experimental studies have not detected This exclusion was chosen because our intent was to study direct actions of serum phosphate on FGF23. Rare heritable FGF23 under normal conditions, because the strong influence disorders are characterized by dysregulated FGF23 metabolism, of kidney disease may overwhelm potentially more subtle in- yet only two (autosomal dominant hypophosphatemic rickets fluences of individual SNPs on circulating FGF23. The indi- and familial hyperphosphatemic tumoral calcinosis) involve di- vidual studies were approved by the local research ethics 13–15 rect mutation of the FGF23 gene. Mutations in other bone committees, and informed consent was obtained from all metabolism genes cause familial tumoral calcinosis (GALNT3, participants. klotho), X-linked phosphatemic rickets (PHEX), and autosomal recessive hypophosphatemic rickets (DMP1, FAM20C, Measurement of FGF23 ENPP1), that subsequently alter FGF23 metabolism through Circulating FGF23 concentrations were measured by ELISA mechanisms that are not completely understood.16–23 We (Kainos Laboratories, Inc., Tokyo, Japan), which detects the performed a meta-analysis of genome-wide association studies full-length, biologically intact FGF23 molecule via midmole- of circulating FGF23 concentrations among 16,624 individuals cule and distal epitopes, in the ARIC, Indiana Sisters, MrOS of European ancestry from seven cohort studies, and replicated GBG and Malmo, and MESA cohorts. In the CHS and OPRA in 4443 individuals of African ancestry from three cohorts, to cohorts, FGF23 was measured using a C-terminal ELISA kit investigate the role of common genetic variants on this (Immutopics, San Clemente, CA). hormone.

Genotyping and Quality Control METHODS Genome-wide genotyping and imputation with reference to the 1000 Genomes Phase 3 genotypes were performed inde- Study Populations pendently in each cohort. Each cohort applied sample and Seven cohorts contributed to the meta-analysis, by providing SNP-based quality control measures for missingness, – study-specific genome-wide analyses of FGF23 concentra- minor allele frequency, and Hardy Weinberg equilibrium tions, for a total of 16,624 individuals of European ancestry (Supplemental Table 1). Poorly imputed SNPs were excluded 2, , fi (Table 1). Contributing studies included the Atherosclerosis if R 0.3 or proper-info was 0.4. Population strati cation and Risk in Communities Study (ARIC; number of individuals of relatedness were assessed using the ancestry principal compo- European ancestry, n=8594),24 the Indiana Sisters Study nents as previously described.31 All cohorts used EIGENSTRAT (n=1128),25 Osteoporotic Fractures in Men Study–Goteborg for principal components of ancestry computation. (MrOS GBG, n=937),26 the Multi-Ethnic Study of Atheroscle- rosis (MESA; number of individuals of European ancestry, Statistical Analyses n=2163),27 MrOS Malmo26 (n=894), the Osteoporosis Pro- Each SNP was tested for association with natural log– spective Risk Assessment Study (OPRA, n=920),28 and the transformed FGF23 using linear regression in two additive ge- Cardiovascular Health Study (CHS; number of individuals netic models.32 Model 1 was adjusted for age, sex, and the first of European ancestry, n=1988).29 Detailed information on ten principal components of ancestry. the study cohorts and methods is provided in Supplemental FGF23 concentrations are strongly associated with eGFR Table 1 and Table 1. and body mass index (BMI). To detect FGF23 loci indepen- We calculated the eGFR using the creatinine-based CKD dent of these pathways and to diminish associations with Epidemiology Collaboration equation equation,30 and effects modulated through these factors, model 2 added

2584 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 2583–2592, 2018 www.jasn.org CLINICAL EPIDEMIOLOGY

Table 1. Cohort and study participant (European descent) characteristics Characteristic ARIC CHS Indiana MESA MrOS GBG MrOS Malmo OPRA Individuals of European descent 8594 1988 1128 2163 937 894 920 Age, yr 57.1 (5.7) 78.0 (4.4) 36.4 (8.5) 62.6 (10.3) 72.3 (5.7) 73.2 (5.8) 75.2 (0.1) Men, n (%) 3962 (46.1) 754 (38.3) 0 (0) 1025 (47.3) 937 (100) 894 (100) 0 (0) BMI, kg/m2 27.3 (5.0) 26.6 (4.5) 26.3 (6.0) 27.7 (5.1) 27.5 (3.7) 27.5 (3.7) 26.2 (4.1) eGFR (CKD-EPI), ml/min per 1.73 m2 87.8 (13.2) 72.1 (17.7) 103.5 (7.4) 76.0 (17.1) 71.6 (19.3) 68.8 (19.3) 68.0 (12.6) FGF23, pg/ml (intact) 45.5 (17.3) — 39.4 (17.4) 41.9 (17.6) 46.2 (22.5) 55.9 (27.7) — FGF23, pg/ml (C-terminal) — 12.9 (21.3) ——— —8.3 (9.1) Data are mean (SD) or number (%), as appropriate. Indiana, Indiana Sisters Cohort; GBG, Goteburg; CKD-EPI, CKD Epidemiology Collaboration equation; —,not available. adjustment for BMI, and both linear and quadratic terms for analyses were performed using the same quality control and the eGFR.30 imputation exclusions, and meta-analysis was performed us- Genomiccontrol parameterswereestimatedforeachcohort ing METAL.34 and appropriate genomic control correction was applied to input statistics before performing meta-analysis to correct Sensitivity Analyses for residual cryptic relatedness or population stratification.33 A sensitivity analysis was conducted in which we excluded the There was little evidence for population stratification at study CHS and OPRA GWAS results, which were on the basis of level (median genomic inflation factor, l=1.03) or meta- C-terminal FGF23 measurements, in order to evaluate whether analysis level (l=1.007). A fixed-effects inverse-variance their inclusion had altered the findings. weighted meta-analysis using METALwas performed across cohorts on the b coefficient/SEM from each cohort.34 Secondarily, eQTL and Functionality Prediction random-effects DerSimonian and Laird models were performed Functional annotations were assigned to all SNPs from the 26 using Stata 15.1 (Statacorp, College Station, TX).35 meta-analysis with P,5310 using ANNOVAR software.40 Genetic differentiation was estimated using the Weir un- We used the MetaXcan software to identify genes whose ex- biased estimator of the fixation index, calculated using the pression levels were significantly associated with circulating variance in allele frequencies among European and African FGF23.41,42 MetaXcan uses the FGF23 GWAS summary re- ancestry samples from the 1000 Genomes and standardized sults and gene expression levels predicted by genetic variants according to the mean allele frequency in the combined in a library of tissues from the Genotype-Tissue Expression sample.36 (GTEx) project to impute the genetic component of gene ex- We examined the consistency of the magnitude and direc- pression in different tissues (thereby eliminating the need to tion of associations across individual studies using forest plots. directly measure gene expression levels) and to correlate the The Locuszoom tool was used to make regional association imputed gene expressions with the phenotype of interest.43,44 plots,37 and Manhattan and QQ-plots were plotted with the For the purposes of this study, we used covariance matrices qqman R package.38 built for ten relevant tissues from GTEx (i.e.,wholeblood cells, transformed fibroblast cells, subcutaneous adipose, skel- Proportion of Phenotypic Variance Explained etal muscle, liver, tibial nerve, left ventricle of the heart, cor- The proportion of variance (PVE) in circulating FGF23 levels onary artery, aortic artery, and atrial appendage). Of note, explained byeach top novel locus, jointly across all cohorts, was reliable data from kidney and bone tissue are not currently estimated as: available in GTEx. MetaXcan is an extension of PredixCan,39 a gene-based b^2z zð 2 Þ approach that uses GWAS summary results to impute the ge- ¼ 2 MAF 1 MAF PVESNP 2 2b^ zMAFzð1 2 MAFÞþððseðb^ÞÞ2z2NzMAFzð1 2 MAFÞÞ netic component of gene expression in different tissues (thus eliminating the need to directly measure gene expression levels), and correlates the imputed gene expressions with phenotypes of ^ ^ where b, seðbÞ, N, and MAF are the effect size estimate of each interest. minor allele on the relative concentration of FGF23, SEM of the effect size, sample size, and MAF for the SNP, respectively.39 Associations with Other Traits We performed evaluations of SNP associations with publicly Follow-Up in African Ancestry Cohorts available results generated from consortia investigating other SNPs found to be associated with FGF23 concentration in traits. Specifically, we evaluated SNPs associated with eGFR from individuals of European ancestry were evaluated for replication CKD-GEN,45 coronary artery disease in CARDIOGRAM,46 among individuals of African ancestry from three cohorts: parathyroid hormone,47 and bone mineral density from the ARIC (n=2464), MESA (n=1510), and CHS (n=469). These GEFOS.48 In total, we performed 20 tests, corresponding to

J Am Soc Nephrol 29: 2583–2592, 2018 Genetic Variants Associated with FGF23 Levels 2585 CLINICAL EPIDEMIOLOGY www.jasn.org

Figure 1. Manhattan plot of SNPs for FGF23. SNPs are plotted on the x axis according to their position on each chromosome against association with ln(FGF23) on the y axis (shown as –log10 P value). four traits tested for association against the five SNPs. Signifi- intergenic SNP lies 38 kb upstream of CYP24A1 (cytochrome cance was evaluated at the Bonferroni-corrected level of P450, family 24, subfamily A, polypeptide 1). Each additional 0.0025. copy of the rs17216707 T allele was associated with 5.4% higher FGF23 concentration, after adjustment for age, sex, and the first ten principal components of ancestry (model 1). RESULTS The CYP24A1 gene encodes the major enzyme responsible for catabolizing calcitriol to water-soluble 1,24,25-trihydroxyvita- The SNP-based meta-analysis identified 192 SNPs associated min D (1,24,25(OH)3D) and 25(OH)D to 25-hydroxyvitamin with circulating FGF23 at genome-wide significance level D (25(OH)D) to 24,25-dihydroxyvitamin D (24,25(OH)2D) for 2 (P,5310 8). These SNPs were located in five genomic re- excretion.49 gions, 5q35.3, 9q21.11, 9q34.2, 16q23.2, and 20q13.2 (Figure 1). rs2769071, a variant on chromosome 9q34.2, intronic in the The top SNP in each region and genes contained in the region ABO (ABO blood group transferase gene), was also associated 2 2 were 20q13.2, rs17216707 (P=3.0310 24; CYP24A1); 9q34.2, with FGF23 concentrations (P=6.1310 17; Supplemental 2 rs2769071 (P=6.1310 17; ABO); 5q35.3, rs11741640 Figure 2B, Table 2). Every additional minor allele at the locus 2 2 (P=1.6310 16; RGS14); 9q21.11, rs17479566 (P=2.0310 9; was associated with 3.7% higher circulating FGF23 concen- 2 LINC01506); and 16q23.2, rs9925837 (P=5.1310 9; trations. The association did not remain statistically signifi- LINC01229). In aggregate, the top five loci explained 3% of cant after adjustment for BMI, eGFR, and eGFR squared 2 the variability in circulating FGF23. Inspection of the quan- (P=3.0310 5). tile-quantile plot suggested an excess of association signals be- Within the 5q35.3 region, the variant most strongly asso- yond those expected (Supplemental Figure 1). All 341 SNPs ciated with circulating FGF23 concentrations was rs11741640 2 from the meta-analysis with a P value less than a suggestive (P=1.6310 16; Table 2). This SNP is intronic within RGS14 2 significance threshold of P,5310 6 are detailed in Supple- (regulator of G-protein signaling 14) and is in linkage disequi- mental Table 2. librium (LD) with SNPs in adjacent genes, SLC34A1 (solute The top SNP in the 20q13.2 region, having the strongest carrier family 34 [type 2 sodium/phosphate cotransporter], association with FGF23 concentration, was rs17216707 member 1, responsible for phosphate reabsorption in the 2 (P=3.0310 24;SupplementalFigure2A,Table2).This proximal tubule) (e.g., rs12659266, R2=0.78), and FGFR4

2586 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 2583–2592, 2018 www.jasn.org CLINICAL EPIDEMIOLOGY

Table 2. Associations of top single nucleotide polymorphisms with ln-transformed FGF23 concentrations c FGF23 FGF23 Model 1 Model 2 Nearest Other n n SNP a Chr Position Increasing Increasing ( =16,624) ( =15,496) Gene Allele b Allele Allele Frequency bd (SEM) P Value b (SEM) P Value 2 2 rs17216707 CYP24A1 20 52732362 T C 0.80 0.054 (0.005) 3.0310 24 0.044 (0.005) 1.5310 21 2 2 rs2769071 ABO 9 136145974 G A 0.37 0.037 (0.005) 6.1310 17 0.031 (0.008) 3.0310 5 2 2 rs11741640 RGS14 5 176792743 G A 0.73 0.039 (0.005) 1.6310 16 0.038 (0.005) 2.8310 16 2 2 rs17479566 LINC01506 9 71198013 T C 0.22 0.031 (0.005) 2.0310 9 0.039 (0.009) 6.8310 6 2 2 rs9925837 LINC01229 16 79927303 G A 0.13 0.035 (0.006) 5.1310 9 0.035 (0.006) 2.1310 9 Only the top SNP from each region is shown. Model 1 includes age, sex, and ﬁrst ten principal components of ancestry. SNP, single nucleotide polymorphism; Chr, chromosome. aNearest gene by physical distance to the lead SNP. bAllele frequency data from 1000 Genomes Phase 1 genotype data. cResults for model 2 were not available for the Indiana Sisters Study. db-estimates are interpreted as the relative difference in FGF23 concentration per minor allele; e.g., 0.051 is a 5.1% higher FGF23 concentration per additional allele.

(fibroblast growth factor receptor 4) (e.g., rs244707, R2=0.43) associated (P,0.05) with FGF23 concentrations (Table 3). (Supplemental Figure 2C).11,50,51 Rs17216707, rs11741640, and rs9925837 showed the most Other top SNPs significantly associated with FGF23 included allelic differentiation between individuals of African and rs17479566,anSNP inthe 9q21.11 region,residing intergenically European ancestry. between LINC01506 (long intergenic non–protein coding RNA 1506) and PIP5K1B (phosphatidylinositol-4-phosphate Sensitivity Analysis 5-kinase type 1 b; Supplemental Figure 2D), and rs9925837, We repeated our meta-analysis after excluding the CHS and an intergenic SNP positioned between LINC01229 (long inter- OPRA, in which circulating concentrations of C-terminal genic non–protein coding RNA 1229) and LOC102724084 (un- FGF23 were measured. This meta-analysis identified 161 characterized noncoding RNA) in the 16q23.2 region SNPs associated with circulating intact FGF23 at genome- 2 2 (P=5.1310 9; Supplemental Figure 2E, Table 2). wide significance level (P,5310 8), located in four genomic The primary regression coefficients and interpretation of regions, 5q35.3, 9q21.11, 9q34.2, and 20q13.2. The top SNP in our results were not affected by further adjustment for BMI, each region and genes contained in the region were 20q13.2, 2 eGFR, and eGFR squared (model 2) for rs17216707, rs17216707 (b [SEM]=0.057 [0.006]; P=1.0310 25; rs11741640, or rs9925837 (Supplemental Figure 3). However, CYP24A1); 9q34.2, rs2769071 (b [SEM]=0.037 [0.005]; 2 the P values for SNPs rs2769071 and rs17479566 were atten- P=7.1310 16; ABO); 5q35.3, rs11741640 (b [SEM]=0.038 2 uated by factors of 1012 and 103, respectively (Table 2). [0.005]; P=1.6310 14; RGS14); and 9q21.11, rs17479566 2 (b [SEM]=0.032 [0.005]; P=3.8310 9; LINC01506) (Supple- Replication in Individuals of African Ancestry mental Table 3). Although the association at rs9925837 did not In populations of African ancestry, the effect estimates for the meet genome-wide significance, the magnitude of the effect five top SNPs were in the same direction as in individuals of was not meaningfully different after exclusion of CHS and 2 European ancestry and one SNP (rs9925837) was nominally OPRA (b [SEM]=0.032 [0.006], P=2.2310 7).

Table 3. Associations of top single nucleotide polymorphisms with ln-transformed FGF23 concentrations among individuals of African ancestry (n=4443) FGF23 Other FGF23 Increasing SNP Nearest Gene Chr Position Increasing bb (SEM) P Value Fstc Allele Allele Frequencya Allele rs17216707 CYP24A1 20 52732362 T C 0.97 0.044 (0.037) 0.24 0.162 rs2769071 ABO 9 136145974 G A 0.48 0.010 (0.012) 0.41 0.009 rs11741640 RGS14 5 176792743 G A 0.96 0.008 (0.027) 0.78 0.195 rs17479566 LINC01506 9 71198013 T C 0.93 0.022 (0.020) 0.29 0.087 rs9925837 LINC01229 16 79927303 G A 0.28 0.029 (0.015) 0.05 0.232 Only the top SNP from each region is shown. SNP, single nucleotide polymorphism; Chr, chromosome. aAllele frequency data from 1000 Genome Phase 1 genotype data among African samples. bEffect estimate adjusted for age, sex, and ﬁrst ten principal components of ancestry. b-estimates are interpreted as the relative difference in FGF23 concentration per minor allele; e.g., 0.038 is a 3.8% higher FGF23 concentration per additional allele. cFst is an estimate of genetic differentiation calculated using the variance in allele frequencies among European- and African-ancestry samples fromthe1000 Genomes and standardized according to the mean allele frequency in the combined sample.36

J Am Soc Nephrol 29: 2583–2592, 2018 Genetic Variants Associated with FGF23 Levels 2587 CLINICAL EPIDEMIOLOGY www.jasn.org

15 12 6 4 Functional Evaluation 2 2 2 2 28

) ﬁ , 3 10 10 10 10 Ofthe192SNPsthatmetgenome-widesigni cance(P 5 10 ) 2 Value 3 3 3 3 in the primary analysis, 98 were in intergenic regions, 84 P overlapped introns, one was exonic (rs5030873, for SLC34A1), d one overlapped a 39 untranslated region (rs3812035, for

eGFR SLC34A1), four overlapped the 1-kb region upstream of a transcription start site (all for RGS14), and four overlapped (SEM) (ml/min per 1.73 m

48 the 1-kb region downstream of a transcription end site (three b 0.97% (0.13) 5.4 for RGS14 and one for LMAN2) (Supplemental Table 2). Anal- +0.01% (0.10)+0.75% (0.11)+0.56% (0.12) 4.5 0.91 1.5 2 ysis with MetaXcan to evaluate the association with genetically predicted gene expression implicated RGS14 as the most sig- 25 4 24 5

2 2 2 2 ﬁ ﬁ

c ni cant genetic result and revealed a signi cant association of 10 10 10 10 Value 3 3 3 3 FGF23 with increased expression at that gene in multiple tis- P sues (Supplemental Table 4). 45 Osteoporosis Consortium. y Artery Disease Genome Wide Replication and Meta- Assessment for Association of SNPs with Related Traits Each of the top SNPs was associated with parathyroid hormone (% difference) concentration; four of the five were significantly associated at (SEM) Parathyroid Hormone b the Bonferroni-corrected P value threshold of 0.003 (Table 4). 1.7% (0.047) 5.1 3.1% (0.031) 3.4 5.4% (0.053) 4.7 – +1.2% (0.044) 6.3 2 2 We also observed associations of four of the five SNPs with eGFR, and of rs2769071 with coronary artery disease and bone

8 mineral density. At this locus, the FGF23 increasing allele was 2

10 associated with 4.5% greater odds of coronary artery disease 2 2 3

b 6 8 Value (P=3.3310 )andlowerBMD(b=20.0197, P=2.7310 ). P 47 cance was evaluated at the Bonferroni-corrected level of 0.0025. SNP, single nucleotide poly- ﬁ DISCUSSION

In this first reported GWAS of FGF23, we identified five loci,

(SEM) located on chromosomes 5, 9, 16, and 20, that were associated Bone Mineral Density ve SNPs. Signi b ﬁ with variation in FGF23 concentrations. The strongest associ- 0.0035 (0.0047) 0.62 +1.0% (0.055) 0.02 +0.51% (0.14) 1.7 0.0197 (0.0036) 2.7 0.0049 (0.0038) 0.10 0.0036 (0.0044) 0.26 (standardized mean difference) 2 2 2 2 ation was observed for rs17216707, located 37 kb upstream of CYP24A1, encoding the primary catabolic enzyme for calcitriol [1,25(OH) D]. Every additional T allele at this locus was

6 2 2 associated with a 5.1% difference in circulating FGF23. Other 10 a

3 fi Value signi cant associations included SNPs located near the gene P encoding proteins related to the first discovered blood group system, ABO, and genes encoding the regulator of G-protein signaling 14 and the Na/Pi cotransporter solute carrier family Coronary 34 member 1. In replication analyses conducted in blacks, the Artery Disease effect estimates for the top five SNPs were in the same direc- OR (SEM) tion as in the individuals of European ancestry; however, confidence intervals were wider, and the results were not statistically significant. T 0.98 (0.011) 0.10 +0.0066 (0.0041) 0.18 T 0.98 (0.012) 0.13 G 1.05 (0.01)G 3.3 1.00 (0.013) 0.87 G 0.98 (0.010) 0.04

Allele Calcitriol, the activated form of vitamin D, directly pro- motes FGF23 gene transcription via enhancement of the promotor region. The cytochrome p450 enzyme CYP24A1 is critically important for maintaining 1,25(OH)2D concen-

Gene trations within a tight physiologic range and preventing vita- 46 Nearest ABO LINC01506 LINC01229 RGS14 CYP24A1 min D toxicity by catalyzing the conversion of 1,25(OH)2Dto 49 1,24,25(OH)3D for subsequent excretion. We found the Association between replicated SNPs and FGF23-related phenotypes Tallele of rs17216707 to be associated with greater circulating FGF23 concentrations, suggesting that the associated haplo-

SNP type may confer suppressed CYP24A1 activity, with a resul- Estimates from meta-analysis of bone mineral density of theEstimates femoral from neck meta-analysis GWAS of among eGFR 32,961 in individuals 110,517 from individuals 17 from studies 33 within studies the within Genetic the Factors Chronic for Kidney Disease Genetics (CKD-Gen) CKDGen Consortium. Association estimates from GWAS meta-analysis of coronary artery disease among 60,801 cases and 123,504 controls from 48 studies within the Coronar Estimates from meta-analysis of parathyroid hormone GWAS among 29,155 individuals from 13 cohort studies. c d b Analysis Consortium. Table 4. rs2769071 rs17479566 rs9925837 In total, we performed 20 tests, corresponding toa four traits tested for association against the morphism; OR, odds ratio; GWAS, Genome-wide association study. rs11741640 rs17216707 tant increase in 1,25(OH)2D and stimulation of FGF23

2588 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 2583–2592, 2018 www.jasn.org CLINICAL EPIDEMIOLOGY production. The T allele was also strongly associated with sodium phosphate cotransporter (Npt2a, SLC34A1)andfibro- lower eGFR in previous GWASs,52,53 and yet adjustment for blast growth factor receptor 4 (FGFR4). FGF23 acts on the kid- eGFR in the association of the SNP with FGF23 did not alter ney to induce phosphaturia by reducing the expression of Npt2a the magnitude of the association. The association of the in the brush border of the proximal tubular cells, and FGFR4 has T allele of rs17216707 with lower serum concentrations of been proposed as a mediator in the associations of FGF23 with PTH is also consistent with suppressed CYP24A1 activity, de- left ventricular hypertrophy.11 celerated 1,25(OH)2D catabolism, and increased 1,25(OH)2D Strengths of our study include the large and diverse sample, levels. In murine homolog models of autosomal dominant the population-based settings, and the comprehensive set of phosphatemic rickets (ADHR) and X-linked hypophosphate- common genetic variants examined. Potential limitations mic rickets (XLH), high FGF23 states, CYP24A1 expression in include a restriction to commonvariants only, discoveryefforts the kidney is elevated.54,55 Interestingly, when these mouse in an exclusively European ancestry sample, limited African strains were crossed with CYP24A1-null mice, the bone ab- ancestry and cFGF23 samples, and a lack of kidney or bone normalities improved, without improvements in the serum tissue in the gene expression–based association methods. biochemical profile.56 Although the observed variants were associated with small rs2769071 is located within the ABO blood group differences in FGF23 concentrations, a difference in circulating locus on chromosome 9, which encodes alpha 1–3- FGF23 of 5% is a risk equivalent for heart failure in the MESA N-acetylgalactosaminyltransferase, a major determinant of study of 15 ml/min per 1.73 m2 lower eGFR or 5 mm Hg higher ABO blood type. The C allele at rs2769071 is highly correlated systolic BP.4 Further, it is possible that the effects may be more (r2=0.86) with a single base pair insertion/deletion (indel) var- pronounced among individuals with mineral metabolism dis- iant in exon 6 of ABO (rs8176719) which characterizes the turbances, such as those with CKD mineral and bone disorder. human O blood group.57 This SNP is associated with ABO It will be crucial to test these results among individuals with gene expression in most tissues in GTEx. Using available CKD, in whom the biologic implications and potential treat- GWAS data of other phenotypes, we observed a significant ment options are most relevant. The clinical relevance of the association of variation at this locus to be associated with GWAS findings may not be specifically the magnitude of the higher odds of coronary artery disease, lower bone mineral difference in FGF23 concentrations, but the underlying biol- density, and higher parathyroid hormone levels. In previous ogy that is revealed. published GWASs, this SNP has been found to be associated In summary, we demonstrate that common genetic variants with plasma vWf and Factor VII,58 coagulation factors levels,59 are associatedwith circulatingFGF23 in adults from thegeneral and myocardial infarction in patients with angiographic coro- population. Several of the genetic variants are closely linked nary artery disease.60 Recent studies have suggested that serum with enzymes, transporters, and receptors known to be critical iron may be a regulator of FGF23 homeostasis, and ferritin to vitamin D and phosphate homeostasis. Further studies are levels have been shown to differ between blood groups.61–63 needed to confirm the SNPs and target loci and elucidate the The attenuation of the statistical significance of the association biologic role of the implicated genes. Candidate genes may be with adjustment for eGFR and BMI suggests that the contri- explored in more comprehensive metabolic studies of FGF23 bution of genetic risk to FGF23 within the ABO loci may be metabolism and in translational animal models that will shed mediated by the SNP’s effects on these covariates. However, new light on the mechanisms and clinical implications of variation at rs2769071 was not associated with circulating FGF23 in vitamin D and phosphate homeostasis and its eGFR45 in the CKD-GEN GWAS or BMI in the GWAS of treatment. 700,000 individuals from the GIANT Consortium.64 The functional relevance of this polymorphism to mineral metabolism and FGF23 regulation awaits further elucidation. ACKNOWLEDGMENTS rs11741640 is located within RGS14, which encodes a G-protein signal–regulating protein and was the only replicated The authors wish to thank the investigators, staff, and participants of SNP significantly associated with both intact and C-terminal the individual participating studies for their valuable contributions. FGF23. Previous GWASs of serum phosphate and parathy- C.R.-C.’s work is supported by K01DK09019 from the National roid hormone concentrations reported strong associa- Institutes of Diabetes and Digestive and Kidney Diseases (NIDDK). The tions with rs4074995, which is in near-perfect LD with Atherosclerosis Risk in Communities Study is carried out as a collab- rs11741640 (R2=0.93).47,65 In our gene-level expression-based orative study supported by National Heart, Lung, and Blood Institute analyses with MetaXcan, RGS14 was the gene with the strongest (NHLBI) contracts (HHSN268201100005C, HHSN268201100006C, association of genetically determined expression with FGF23 HHSN268201100007C, HHSN268201100008C, HHSN268201100009C, levels. We found that increased expression of RGS14 was asso- HHSN268201100010C, HHSN268201100011C, and HHSN268201100012C), ciated with higher levels of FGF23 across many tissues, including R01HL087641, R01HL59367, R01HL086694, National Human Genome in heart and muscle tissue, lending biologic support to its im- Research Institute grant U01HG004402, and National Institutes of Health plication in mineral metabolism. rs11741640 is also in strong (NIH) contract HHSN268200625226C. FGF23 measurements were LD with multiple SNPs within the gene coding for type 2a supported by R01HL103706. Infrastructure was partly supported by

J Am Soc Nephrol 29: 2583–2592, 2018 Genetic Variants Associated with FGF23 Levels 2589 CLINICAL EPIDEMIOLOGY www.jasn.org grant number UL1RR025005, a component of the NIH and NIH Road- Research Resources, grant UL1RR033176; the National Center for map for Medical Research. The work within the Indiana Sisters cohort was Advancing Translational Sciences, grant UL1TR001881; and the supported by NIH grants R01AG041517, P30AR072581, P01AG018397, NIDDK Diabetes Research Center (DRC) grant DK063491 to the R21AR061078, K23AR057096, UL1RR025761, and M01RR00750 and Southern California Diabetes Endocrinology Research Center. by a donation from the Scottish Rite of Indianapolis Foundation. Gen- Funding for SHARe genotyping was provided by NHLBI contract otyping services were provided by the Center for Inherited Disease Re- N02-HL-64278. Genotyping was performed at Affymetrix (Santa search, which is fully funded through a federal contract from the NIH to Clara, CA) and the Broad Institute of Harvard and MIT (Boston, MA) the Johns Hopkins University (contract HHSN268200782096C). This using the Affymetrix Genome-Wide Human SNP Array 6.0. researchwas supported in part by the Intramural Research Program of the The content is solely the responsibility of the authors and does not NIH, National Library of Medicine. The Osteoporotic Fractures in Men necessarily represent the official views of the NIH. (MrOS) Study is supported by the following NIH institutes: the National Institute on Aging (NIA), the National Institute of Arthritis and Mus- culoskeletal and Skin Diseases, the National Center for Advancing DISCLOSURES Translational Sciences, and the NIH Roadmap for Medical Research None. under the following grant numbers: U01 AG027810, U01 AG042124, U01 AG042139, U01 AG042140, U01 AG042143, U01 AG042145, U01 AG042168, U01 AR066160, and UL1 TR000128. MrOS Sweden is sup- REFERENCES ported by the Swedish Research Council, Läkarutbildningsavtalet/Avtal om läkarutbildning och forskning research grants in Gothenburg, and the 1. Hu MC, Shi M, Zhang J, Pastor J, Nakatani T, Lanske B, et al.: Klotho: King Gustav V and Queen Victoria Frimurarestiftelse Research Founda- A novel phosphaturic substance acting as an autocrine enzyme in the – tion.TheworkwasalsosupportedbytheUKMedicalResearchCouncil renal proximal tubule. FASEB J 24: 3438 3450, 2010 2. Shimada T, Hasegawa H, Yamazaki Y, Muto T, Hino R, Takeuchi Y, et al.: Lifecourse Epidemiology Unit (4050502589). The Osteoporosis Pro- FGF-23 is a potent regulator of vitamin D metabolism and phosphate spective Risk Assessment study was supported by Swedish Research homeostasis. J Bone Miner Res 19: 429–435, 2004 Council grants (K2009-53X-14691-07-3 through K2015-52X-14691- 3. Kurosu H, Ogawa Y, Miyoshi M, Yamamoto M, Nandi A, Rosenblatt KP, 13-4). This Cardiovascular Health Study (CHS) research was supported et al.: Regulation of fibroblast growth factor-23 signaling by klotho. – by NHLBI contracts HHSN268201200036C, HHSN268200800007C, J Biol Chem 281: 6120 6123, 2006 4. Kestenbaum B, Sachs MC, Hoofnagle AN, Siscovick DS, Ix JH, HHSN268200960009C, HHSN268201800001C, N01HC55222, N01HC85079, Robinson-Cohen C, et al.: Fibroblast growth factor-23 and cardiovas- N01HC85080, N01HC85081, N01HC85082, N01HC85083, and cular disease in the general Population: The multi-ethnic study of ath- N01HC85086; and NHLBI grants U01HL080295, R01HL087652, erosclerosis. Circ Heart Fail 7: 409–417, 2014 R01HL085251, R01HL105756, R01HL103612, R01HL120393, and 5. Ix JH, Katz R, Kestenbaum BR, de Boer IH, Chonchol M, Mukamal KJ, U01HL130114, with additional contribution from the National In- et al.: Fibroblast growth factor-23 and death, heart failure, and cardiovascular events in community-living individuals: CHS (Cardiovas- stitute of Neurological Disorders and Stroke. Additional support was cular Health Study). JAmCollCardiol60: 200–207, 2012 provided through R01AG023629 from the NIA. A full list of prin- 6. Isakova T, Wahl P, Vargas GS, Gutiérrez OM, Scialla J, Xie H, et al.: Fi- cipal CHS investigators and institutions can be found at CHS- broblast growth factor 23 is elevated before parathyroid hormone and NHLBI.org. The provision of genotyping data was supported in part phosphate in chronic kidney disease. Kidney Int 79: 1370–1378, 2011 by the National Center for Advancing Translational Sciences, Clin- 7. Gutiérrez OM, Mannstadt M, Isakova T, Rauh-Hain JA, Tamez H, Shah A, et al.: Fibroblast growth factor 23 and mortality among patients ical and Translational Science Institute grant UL1TR001881, and the undergoing hemodialysis. NEnglJMed359: 584–592, 2008 NIDDK Diabetes Research Center (DRC) grant DK063491 to the 8. Kendrick J, Cheung AK, Kaufman JS, Greene T, Roberts WL, Smits G, Southern California Diabetes Endocrinology Research Center. et al.: HOST Investigators: FGF-23 associates with death, cardiovas- Support for FGF23 measurements was provided by American Heart cular events, and initiation of chronic dialysis. J Am Soc Nephrol 22: Association grant 0575021N to J.H.I. Infrastructure for the Cohorts 1913–1922, 2011 for Heart and Aging Research in Genomic Epidemiology Consor- 9. Scialla JJ, Xie H, Rahman M, Anderson AH, Isakova T, Ojo A, et al.: Chronic Renal Insufficiency Cohort (CRIC) Study Investigators: Fibro- tium is supported in part by the NHLBI grant R01HL105756. The blast growth factor-23 and cardiovascular events in CKD. JAmSoc Multi-Ethnic Study of Atherosclerosis (MESA) and the MESA Single Nephrol 25: 349–360, 2014 Nucleotide Polymorphism Health Association Resource (SHARe) 10. Ärnlöv J, Carlsson AC, Sundström J, Ingelsson E, Larsson A, Lind L, project are conducted and supported by the NHLBI in collabora- et al.: Higher fibroblast growth factor-23 increases the risk of all-cause – tion with MESA investigators. Support for MESA is provided by and cardiovascular mortality in the community. Kidney Int 83: 160 166, 2013 contracts HHSN268201500003I, N01-HC-95159, N01-HC-95160, 11. Grabner A, Schramm K, Silswal N, Hendrix M, Yanucil C, Czaya B, et al.: N01-HC-95161, N01-HC-95162, N01-HC-95163, N01-HC-95164, FGF23/FGFR4-mediated left ventricular hypertrophy is reversible. Sci N01-HC-95165, N01-HC-95166, N01-HC-95167, N01-HC-95168, Rep 7: 1993, 2017 N01-HC-95169, UL1-TR-000040, UL1-TR-001079, and UL1- 12. Lutsey PL, Alonso A, Selvin E, Pankow JS, Michos ED, Agarwal SK, et al.: TR-001420. Funding for FGF23 measurements was provided by Fibroblast growth factor-23 and incident coronary heart disease, heart failure, and cardiovascular mortality: The Atherosclerosis Risk in Com- R01HL096875. Support is also provided by grants and contracts munities study. JAmHeartAssoc3: e000936, 2014 R01HL071051, R01HL071205, R01HL071250, R01HL071251, 13. ADHR Consortium: Autosomal dominant hypophosphataemic rickets is R01HL071258, and R01HL071259; by the National Center for associated with mutations in FGF23. Nat Genet 26: 345–348, 2000

2590 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 2583–2592, 2018 www.jasn.org CLINICAL EPIDEMIOLOGY

14. Econs MJ, McEnery PT, Lennon F, Speer MC: Autosomal dominant 34. Willer CJ, Li Y, Abecasis GR: METAL: Fast and efficient meta-analysis of hypophosphatemic rickets is linked to chromosome 12p13. JClinIn- genomewide association scans. Bioinformatics 26: 2190–2191, 2010 vest 100: 2653–2657, 1997 35. DerSimonian R, Laird N: Meta-analysis in clinical trials. Control Clin 15. Larsson T, Yu X, Davis SI, Draman MS, Mooney SD, Cullen MJ, et al.: Trials 7: 177–188, 1986 A novel recessive mutation in fibroblast growth factor-23 causes 36. Weir BS, Cockerham CC: Estimating F-statistics for the analysis of familial tumoral calcinosis. J Clin Endocrinol Metab 90: 2424–2427, 2005 population structure. Evolution 38: 1358–1370, 1984 16. Feng JQ, Ward LM, Liu S, Lu Y, Xie Y, Yuan B, et al.: Loss of DMP1 37. Pruim RJ, Welch RP, Sanna S, Teslovich TM, Chines PS, Gliedt TP, et al.: causes rickets and osteomalacia and identifies a role for osteocytes in LocusZoom: Regional visualization of genome-wide association scan mineral metabolism. Nat Genet 38: 1310–1315, 2006 results. Bioinformatics 26: 2336–2337, 2010 17. Lorenz-Depiereux B, Bastepe M, Benet-Pagès A, Amyere M, 38. Turner SD: qqman: an R package for visualizing GWAS results using Wagenstaller J, Müller-Barth U, et al.: DMP1 mutations in autosomal Q-Q and manhattan plots. Journal of Open Source Software 3: 731, recessive hypophosphatemia implicate a bone matrix protein in the 2018 regulation of phosphate homeostasis. Nat Genet 38: 1248–1250, 2006 39. Shim H, Chasman DI, Smith JD, Mora S, Ridker PM, Nickerson DA, et al.: 18. Econs MJ, Fain PR, Norman M, Speer MC, Pericak-Vance MA, Becker A multivariate genome-wide association analysis of 10 LDL sub- PA, et al.: Flanking markers define the X-linked hypophosphatemic fractions, and their response to statin treatment, in 1868 Caucasians. rickets gene locus. J Bone Miner Res 8: 1149–1152, 1993 PLoS One 10: e0120758, 2015 19. Econs MJ, Francis F, Rowe PS, Speer MC, O’Riordan JL, Lehrach H, 40. Yang H, Wang K: Genomic variant annotation and prioritization with et al.: Dinucleotide repeat polymorphism at the DXS1683 locus. Hum ANNOVAR and wANNOVAR. Nat Protoc 10: 1556–1566, 2015 Mol Genet 3: 680, 1994 41. Barbeira A, Shah KP, Torres JM, Wheeler HE, Torstenson ES, Edwards 20. Francis F, Rowe PS, Econs MJ, See CG, Benham F, O’Riordan JL, et al.: T, et al.: MetaXcan: Summary statistics based gene-level association A YAC contig spanning the hypophosphatemic rickets disease gene method infers accurate PrediXcan results [Internet]. bioRxiv 2016. (HYP) candidate region. Genomics 21: 229–237, 1994 Available at: https://www.biorxiv.org/content/early/2016/03/23/ 21. Rowe PS, Goulding JN, Francis F, Oudet C, Econs MJ, Hanauer A, 045260. Accessed November 10, 2017 et al.: The gene for X-linked hypophosphataemic rickets maps to a 42. Gamazon ER, Wheeler HE, Shah KP, Mozaffari SV, Aquino-Michaels K, 200-300kb region in Xp22.1, and is located on a single YAC Carroll RJ, et al.: GTEx Consortium: A gene-based association method containing a putative vitamin D response element (VDRE). Hum Genet for mapping traits using reference transcriptome data. Nat Genet 47: 97: 345–352, 1996 1091–1098, 2015 22. Francis F, Hennig S, Korn B, Reinhardt R, de Jong P, Poustka A, et al.: 43. Carithers LJ, Ardlie K, Barcus M, Branton PA, Britton A, Buia SA, et al.: A gene (PEX) with homologies to endopeptidases is mutated in pa- GTEx Consortium: A novel approach to high-quality postmortem tissue tients with X-linked hypophosphatemic rickets. The HYP Consortium. procurement: The GTEx project. Biopreserv Biobank 13: 311–319, Nat Genet 11: 130–136, 1995 2015 23. Econs MJ, Rowe PS, Francis F, Barker DF, Speer MC, Norman M, et al.: 44. Melé M, Ferreira PG, Reverter F, DeLuca DS, Monlong J, Sammeth M, Fine structure mapping of the human X-linked hypophosphatemic et al.: GTEx Consortium: Human genomics. The human transcriptome rickets gene locus. J Clin Endocrinol Metab 79: 1351–1354, 1994 across tissues and individuals. Science 348: 660–665, 2015 24. The ARIC Investigators: The Atherosclerosis Risk in Communities 45. Gorski M, van der Most PJ, Teumer A, Chu AY, Li M, Mijatovic V, et al.: (ARIC) Study: Design and objectives. The ARIC investigators. Am J 1000 Genomes-based meta-analysis identifies 10 novel loci for kidney Epidemiol 129: 687–702, 1989 function. Sci Rep 7: 45040, 2017 25. Peacock M, Koller DL, Hui S, Johnston CC, Foroud T, Econs MJ: Peak 46. Nikpay M, Goel A, Won H-H, Hall LM, Willenborg C, Kanoni S, et al.: bone mineral density at the hip is linked to chromosomes 14q and 15q. A comprehensive 1,000 Genomes-based genome-wide association Osteoporos Int 15: 489–496, 2004 meta-analysis of coronary artery disease. Nat Genet 47: 1121–1130, 26. Mellström D, Johnell O, Ljunggren O, Eriksson A-L, Lorentzon M, 2015 Mallmin H, et al.: Free testosterone is an independent predictor of BMD 47. Robinson-Cohen C, Lutsey PL, Kleber ME, Nielson CM, Mitchell BD, Bis and prevalent fractures in elderly men: MrOS Sweden. J Bone Miner JC, et al.: Genetic variants associated with circulating parathyroid Res 21: 529–535, 2006 hormone. J Am Soc Nephrol 28: 1553–1565, 2017 27. Bild DE, Bluemke DA, Burke GL, Detrano R, Diez Roux AV, Folsom AR, 48. Estrada K, Styrkarsdottir U, Evangelou E, Hsu Y-H, Duncan EL, Ntzani et al.: Multi-Ethnic Study of Atherosclerosis: Objectives and design. Am EE, et al.: Genome-wide meta-analysis identifies 56 bone mineral JEpidemiol156: 871–881, 2002 density loci and reveals 14 loci associated with risk of fracture. Nat 28. Gerdhem P, Brändström H, Stiger F, Obrant K, Melhus H, Ljunggren O, Genet 44: 491–501, 2012 et al.: Association of the collagen type 1 (COL1A 1) Sp1 binding site 49. Jones G, Prosser DE, Kaufmann M: 25-Hydroxyvitamin D-24-hydroxy- polymorphism to femoral neck bone mineral density and wrist fracture lase (CYP24A1): Its important role in the degradation of vitamin D. Arch in 1044 elderly Swedish women. Calcif Tissue Int 74: 264–269, 2004 Biochem Biophys 523: 9–18, 2012 29. Fried LP, Borhani NO, Enright P, Furberg CD, Gardin JM, Kronmal RA, 50. Tenenhouse HS: Regulation of phosphorus homeostasis by the type iia et al.: The Cardiovascular Health Study: Design and rationale. Ann na/phosphate cotransporter. Annu Rev Nutr 25: 197–214, 2005 Epidemiol 1: 263–276, 1991 51. Branch MR, Hepler JR: Endogenous RGS14 is a cytoplasmic-nuclear 30. Inker LA, Schmid CH, Tighiouart H, Eckfeldt JH, Feldman HI, Greene T, shuttling protein that localizes to juxtanuclear membranes and chro- et al.: CKD-EPI Investigators: Estimating glomerular filtration rate from matin-rich regions of the nucleus. PLoS One 12: e0184497, 2017 serum creatinine and cystatin C. NEnglJMed367: 20–29, 2012 52. Pattaro C, Teumer A, Gorski M, Chu AY, Li M, Mijatovic V, et al.: ICBP 31. Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, Reich D: Consortium; AGEN Consortium; CARDIOGRAM; CHARGe-Heart Fail- Principal components analysis corrects for stratification in genome- ure Group; ECHOGen Consortium: Genetic associations at 53 loci wide association studies. Nat Genet 38: 904–909, 2006 highlight cell types and biological pathways relevant for kidney func- 32. Zheng J, Li Y, Abecasis GR, Scheet P: A comparison of approaches tion. Nat Commun 7: 10023, 2016 to account for uncertainty in analysis of imputed genotypes. Genet 53. Mahajan A, Rodan AR, Le TH, Gaulton KJ, Haessler J, Stilp AM, et al.: Epidemiol 35: 102–110, 2011 SUMMIT Consortium; BioBank Japan Project: Trans-ethnic fine map- 33. Devlin B, Roeder K: Genomic control for association studies. Biometrics ping highlights kidney-function genes linked to salt sensitivity. Am J 55: 997–1004, 1999 Hum Genet 99: 636–646, 2016

J Am Soc Nephrol 29: 2583–2592, 2018 Genetic Variants Associated with FGF23 Levels 2591 CLINICAL EPIDEMIOLOGY www.jasn.org

54. Bai X-Y, Miao D, Goltzman D, Karaplis AC: The autosomal dominant Control Consortium: Identification of ADAMTS7 as a novel locus for hypophosphatemic rickets R176Q mutation in fibroblast growth factor coronary atherosclerosis and association of ABO with myocardial in- 23 resists proteolytic cleavage and enhances in vivo biological potency. farction in the presence of coronary atherosclerosis: Two genome-wide JBiolChem278: 9843–9849, 2003 association studies. Lancet 377: 383–392, 2011 55. Roy S, Martel J, Ma S, Tenenhouse HS: Increased renal 25-hydrox- 61. Lewerin C, Ljunggren Ö, Nilsson-Ehle H, Karlsson MK, Herlitz H, yvitamin D3-24-hydroxylase messenger ribonucleic acid and immu- Lorentzon M, et al.: Low serum iron is associated with high serum intact noreactive protein in phosphate-deprived Hyp mice: A mechanism for FGF23 in elderly men: The Swedish MrOS study. Bone 98: 1–8, 2017 accelerated 1,25-dihydroxyvitamin D3 catabolism in X-linked hypo- 62. Imel EA, Liu Z, McQueen AK, Acton D, Acton A, Padgett LR, et al.: phosphatemic rickets. Endocrinology 134: 1761–1767, 1994 Serum fibroblast growth factor 23, serum iron and bone mineral density 56. Bai X, Miao D, Xiao S, Qiu D, St-Arnaud R, Petkovich M, et al.: CYP24 in premenopausal women. Bone 86: 98–105, 2016 inhibition as a therapeutic target in FGF23-mediated renal phosphate 63. Rigas AS, Berkfors AA, Pedersen OB, Sørensen E, Nielsen KR, Larsen wasting disorders. JClinInvest126: 667–680, 2016 MH, et al.: Reduced ferritin levels in individuals with non-O blood 57. Yamamoto F, Clausen H, White T, Marken J, Hakomori S: Molecular group: Results from the Danish Blood Donor Study. Transfusion 57: genetic basis of the histo-blood group ABO system. Nature 345: 2914–2919, 2017 229–233, 1990 64. Yengo L, Sidorenko J, Kemper KE, Zheng Z, Wood AR, Weedon MN, 58. Smith NL, Chen M-H, Dehghan A, Strachan DP, Basu S, Soranzo N, et al.: GIANT Consortium: Meta-analysis of genome-wide association et al.: Wellcome Trust Case Control Consortium: Novel associations of studies for height and body mass index in ~700000 individuals of Eu- multiple genetic loci with plasma levels of factor VII, factor VIII, and von ropean ancestry [published online ahead of print August 16, 2018]. Willebrand factor: The CHARGE (Cohorts for Heart and Aging Research Hum Mol Genet doi: 10.1093/hmg/ddy271 in Genome Epidemiology) Consortium. Circulation 121: 1382–1392, 65. Kestenbaum B, Glazer NL, Köttgen A, Felix JF, Hwang SJ, Liu Y, et al.: 2010 Common genetic variants associate with serum phosphorus concen- 59. Desch KC, Ozel AB, Siemieniak D, Kalish Y, Shavit JA, Thornburg CD, tration. JAmSocNephrol7: 1223–1232, 2010 et al.: Linkage analysis identifies a locus for plasma von Willebrand factor undetected by genome-wide association. Proc Natl Acad Sci U S A 110: 588–593, 2013 60. Reilly MP, Li M, He J, Ferguson JF, Stylianou IM, Mehta NN, et al.: This article contains supplemental material online at http://jasn.asnjournals. Myocardial Infarction Genetics Consortium; Wellcome Trust Case org/lookup/suppl/doi:10.1681/ASN.2018020192/-/DCSupplemental.

AFFILIATIONS

1Division of Nephrology, Department of Medicine, Vanderbilt University Medical Center, Nashville, Tennessee; 2Cardiovascular Health Research Unit, Departments of Biostatistics and Medicine, 18Cardiovascular Health Research Unit, Departments of Epidemiology, Health Services and Medicine, and 20Kidney Research Institute, Division of Nephrology, Department of Medicine, University of Washington, Seattle, Washington; Departments of 3Medical and Molecular Genetics and 4Medicine, Indiana University, Indianapolis, Indiana; 5Division of Cardiology, Johns Hopkins School of Medicine, Baltimore, Maryland; 6Clinical and Molecular Osteoporosis Research Unit, Department of Clinical Science Malmö, Lund University, Malmö, Sweden; 7Department of Orthopaedics, Skåne University Hospital, Malmö, Sweden; 8The Institute for Translational Genomics and Population Sciences, Department of Pediatrics, Los Angeles Biomedical Research Institute at Harbor– University of California, Los Angeles Medical Center, Torrance, California; 9Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, and 10Welch Center for Prevention, Epidemiology, and Clinical Research, Johns Hopkins University, Baltimore, Maryland; 11Department of Medicine, Bioinformatics Core Facility and 12Centre for Bone and Arthritis Research, Department of Internal Medicine and Clinical Nutrition, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden; 13Division of Nephrology, Department of Medicine, and Duke Clinical Research Institute, Duke University School of Medicine, Durham, North Carolina; 14Department of Medical Sciences, Endocrinology and Mineral Metabolism, Uppsala University, Uppsala, Sweden; 15Division of Nephrology-Hypertension, Department of Medicine, University of California San Diego, San Diego, California; 16Nephrology Section, Veterans Affairs San Diego Healthcare System, San Diego, California; 17Division of Epidemiology and Community Health, School of Public Health, University of Minnesota, Minneapolis, Minnesota; 19Kaiser Permanente Washington Health Research Institute, Seattle, Washington

2592 Journal of the American Society of Nephrology J Am Soc Nephrol 29: 2583–2592, 2018 Table of Contents

Table 1. Genotyping, quality control, imputation, and analysis methods in the participating studies ..... 2 Figure 1. QQ-Plot ...... 3 Figure 2a. Regional Association Plot, rs17216707 ...... 4 Figure 2b. Regional Association Plot, rs2769071...... 5 Figure 2c. Regional Association Plot, rs11741640 ...... 6 Figure 2d. Regional Association Plot, rs17479566...... 7 Figure 2e. Regional Association Plot, rs9925837 ...... 8 Figure 3. Manhattan Plot, Model 2, European Ancestry ...... 9 Figure 4a. Forest plot of beta-coefficients for rs17216707 from individual studies (Model 1) ...... 10 Figure 4b. Forest plot of beta-coefficients for rs2769071 from individual studies (Model 1) ...... 10 Figure 4c. Forest plot of beta-coefficients for rs11741640 from individual studies (Model 1) ...... 11 Figure 4d. Forest plot of beta-coefficients for rs17479566 from individual studies (Model 1) ...... 11 Figure 4e. Forest plot of beta-coefficients for rs9925837 from individual studies (Model 1) ...... 12 Supplementary Table 2. Annotated SNPs demonstrating suggestive significance ...... 13 Supplementary Table 3. Associations of top SNPs with ln-transformed intact FGF23 concentration .. 25 Supplementary Table 4. Genetically predicted gene expression results ...... 26

Table 1. Genotyping, quality control, imputation, and analysis methods in the participating studies

Genomic SNP Inclusion Criteria control parameter Study Genotyping Platform/Chip SNP Call MAF HWE rate p-value

ARIC Affymetrix 6.0 ≥0.95 ≥0.01 ≥10-4 1.036 CHS European ancestry: Illumina ≥0.97 ≥0.01 ≥10-4 1.009 370CNV BeadChip merged with IBC Illimina iSELECT African ancestry: Illumina Human Omni1-Quad_v1 Indiana Illumina Quad 660 BeadChip ≥0.95 ≥0.03 ≥10-4 1.175 MESA Affymetrix 6.0 ≥0.95 ≥0.01 ≥10-4 1.020 MrOS (GBG and Malmo) Illumina Human Omni1-Quad_v1 ≥0.98 ≥0.01 ≥10-4 1.023 OPRA Affymetrix 6.0 ≥0.97 ≥0.01 ≥10-4 1.014 *CHS additionally excluded for heterozygote frequency=0, ≥1 duplicate error, or Mendelian inconsistency in HapMap controls. All cohorts used EIGENSTRAT for principal components of ancestry computation.

Figure 1. QQ-Plot (Genomic inflation factor λ=1.007)

Figure 2a. Regional Association Plot, rs17216707

Figure 2b. Regional Association Plot, rs2769071

Figure 2c. Regional Association Plot, rs11741640

Figure 2d. Regional Association Plot, rs17479566

Figure 2e. Regional Association Plot, rs9925837

Figure 3. Manhattan Plot, Model 2, European Ancestry

Figure 4a. Forest plot of beta-coefficients for rs17216707 from individual studies (Model 1)

Figure 4b. Forest plot of beta-coefficients for rs2769071 from individual studies (Model 1)

Figure 4c. Forest plot of beta-coefficients for rs11741640 from individual studies (Model 1)

Figure 4d. Forest plot of beta-coefficients for rs17479566 from individual studies (Model 1)

Figure 4e. Forest plot of beta-coefficients for rs9925837 from individual studies (Model 1)

Supplementary Table 2. Annotated SNPs demonstrating suggestive significance (P<5 x 10-6) (Model 1) MarkerName Effect Other Effect Allele Effect SE P.value.FE P.value.RE Chr Position Overlapped Nearest Nearest Allele Allele Frequency Gene Upstream Gene Downstream Gene rs17216707 t c 0.7997 0.0542 0.005 3.00E-24 1.10E-11 chr20 52732362 BCAS1 CYP24A1 rs6127099 a t 0.7207 0.0477 0.005 4.16E-22 5.80E-16 chr20 52731402 BCAS1 CYP24A1 rs2769071 a g 0.6491 -0.037 0.005 6.08E-17 4.79E-07 chr9 136145974 ABO rs527210 t c 0.3428 0.038 0.005 6.38E-17 3.49E-07 chr9 136146431 ABO rs687289 a g 0.3434 0.0379 0.005 7.34E-17 1.42E-06 chr9 136137106 ABO rs543040 a t 0.6583 -0.038 0.005 1.06E-16 6.00E-07 chr9 136143000 ABO rs8121940 c g 0.7902 0.0457 0.006 1.13E-16 4.80E-19 chr20 52742306 BCAS1 CYP24A1 rs35194449 t c 0.2095 -0.046 0.006 1.18E-16 4.84E-19 chr20 52742047 BCAS1 CYP24A1 rs505922 t c 0.6516 -0.037 0.005 1.43E-16 7.37E-07 chr9 136149229 ABO rs8176663 t c 0.6501 -0.037 0.005 1.56E-16 6.74E-07 chr9 136144427 ABO rs11741640 a g 0.2684 -0.039 0.005 1.59E-16 6.10E-17 chr5 176792743 RGS14 rs17217119 a g 0.79 0.0451 0.006 1.63E-16 3.72E-17 chr20 52742590 BCAS1 CYP24A1 rs6013897 a t 0.2102 -0.045 0.006 1.66E-16 7.48E-19 chr20 52742479 BCAS1 CYP24A1 rs529565 t c 0.6519 -0.037 0.005 1.69E-16 7.25E-07 chr9 136149500 ABO rs491626 t c 0.3484 0.037 0.005 1.76E-16 7.29E-07 chr9 136144873 ABO rs514659 a c 0.6503 -0.037 0.005 1.77E-16 7.00E-07 chr9 136142203 ABO rs492488 a g 0.3504 0.0369 0.005 1.86E-16 7.36E-07 chr9 136144960 ABO rs545971 t c 0.3497 0.0369 0.005 1.88E-16 6.94E-07 chr9 136143372 ABO rs554833 t c 0.3499 0.0369 0.005 1.89E-16 7.52E-07 chr9 136147160 ABO rs674302 a t 0.3498 0.0369 0.005 1.94E-16 7.76E-07 chr9 136146664 ABO rs676457 a t 0.6504 -0.037 0.005 1.98E-16 7.79E-07 chr9 136146227 ABO rs493246 a g 0.3507 0.0368 0.005 1.99E-16 6.82E-07 chr9 136144994 ABO rs495203 t c 0.3505 0.0368 0.005 2.12E-16 7.58E-07 chr9 136145240 ABO rs582094 a t 0.6502 -0.037 0.005 2.22E-16 7.01E-07 chr9 136145484 ABO

13 rs582118 a g 0.6502 -0.037 0.005 2.22E-16 7.01E-07 chr9 136145471 ABO rs687621 a g 0.6492 -0.037 0.005 2.55E-16 1.09E-06 chr9 136137065 ABO rs11746443 a g 0.2675 -0.039 0.005 6.67E-16 6.16E-11 chr5 176798306 RGS14 rs612169 a g 0.6509 -0.036 0.005 9.10E-16 1.16E-06 chr9 136143442 ABO rs11748297 a g 0.2758 -0.038 0.005 9.96E-16 3.74E-16 chr5 176800361 RGS14 SLC34A1 rs4075958 a g 0.2687 -0.038 0.005 1.04E-15 2.91E-15 chr5 176784512 LMAN2 RGS14 rs11748165 t c 0.2788 -0.038 0.005 1.13E-15 4.09E-16 chr5 176800054 RGS14 SLC34A1 rs2870308 a c 0.2918 -0.039 0.005 1.96E-15 1.37E-17 chr20 52727953 BCAS1 CYP24A1 rs677355 a g 0.3407 0.0362 0.005 2.15E-15 8.82E-07 chr9 136146046 ABO rs4074995 a g 0.2771 -0.037 0.005 2.31E-15 1.35E-09 chr5 176797343 RGS14 rs13153019 t c 0.7444 0.0392 0.005 2.33E-15 8.73E-16 chr5 176782218 LMAN2 RGS14 rs209956 a g 0.7208 0.0373 0.005 2.78E-15 2.65E-16 chr20 52722856 BCAS1 CYP24A1 rs209954 a t 0.7204 0.0373 0.005 3.10E-15 2.68E-16 chr20 52722943 BCAS1 CYP24A1 rs209955 t c 0.2796 -0.037 0.005 3.14E-15 6.45E-17 chr20 52722871 BCAS1 CYP24A1 rs10051765 t c 0.6765 0.0378 0.005 3.19E-15 8.61E-12 chr5 176799992 RGS14 SLC34A1 rs56235845 t g 0.6884 0.0383 0.005 3.90E-15 1.25E-15 chr5 176798040 RGS14 rs676996 t g 0.6589 -0.036 0.005 3.96E-15 1.03E-06 chr9 136146077 ABO rs209957 a g 0.72 0.0369 0.005 4.93E-15 1.09E-16 chr20 52721095 BCAS1 CYP24A1 rs576125 a g 0.3187 0.0371 0.005 7.29E-15 2.32E-06 chr9 136144309 ABO rs597974 a g 0.6802 -0.037 0.005 9.58E-15 2.93E-06 chr9 136144297 ABO rs643434 a g 0.3711 0.0342 0.005 1.68E-14 5.84E-07 chr9 136142355 ABO rs657152 a c 0.372 0.0341 0.005 1.76E-14 6.04E-07 chr9 136139265 ABO rs597988 a t 0.3202 0.0364 0.005 1.79E-14 2.15E-06 chr9 136144284 ABO rs644234 t g 0.629 -0.034 0.005 1.85E-14 5.44E-07 chr9 136142217 ABO rs544873 a g 0.3707 0.0341 0.005 1.92E-14 5.65E-07 chr9 136143212 ABO rs494242 t c 0.3718 0.034 0.005 2.19E-14 6.35E-07 chr9 136145118 ABO rs12654812 a g 0.3272 -0.036 0.005 5.45E-14 4.52E-10 chr5 176794191 RGS14 rs550057 t c 0.2741 0.0361 0.005 6.51E-14 3.00E-07 chr9 136146597 ABO rs543968 t c 0.6413 -0.034 0.005 7.51E-14 4.78E-07 chr9 136143121 ABO rs613534 a g 0.6412 -0.034 0.005 7.54E-14 4.82E-07 chr9 136143120 ABO

14 rs9411378 a c 0.245 0.0385 0.005 1.29E-13 4.14E-07 chr9 136145425 ABO rs4075959 a g 0.286 -0.035 0.005 1.44E-13 5.51E-14 chr5 176784612 LMAN2 RGS14 rs4976688 t c 0.7138 0.0347 0.005 1.44E-13 5.52E-14 chr5 176784439 LMAN2 RGS14 rs209958 t g 0.2785 -0.035 0.005 1.47E-13 2.44E-14 chr20 52717172 BCAS1 CYP24A1 rs158530 a g 0.2911 -0.034 0.005 3.36E-13 8.08E-14 chr20 52720530 BCAS1 CYP24A1 rs11135015 t c 0.6774 0.0344 0.005 4.92E-13 1.59E-13 chr5 176792557 RGS14 rs6556313 a g 0.676 0.0344 0.005 5.00E-13 1.67E-13 chr5 176792491 RGS14 rs67111717 a g 0.6724 0.034 0.005 1.00E-12 3.37E-13 chr5 176790162 RGS14 rs4976647 a c 0.6745 0.0338 0.005 1.33E-12 4.55E-13 chr5 176788622 RGS14 rs4976646 t c 0.6729 0.0337 0.005 1.54E-12 1.33E-12 chr5 176788570 RGS14 rs4976689 c g 0.6779 0.0331 0.005 2.51E-12 9.31E-13 chr5 176784448 LMAN2 RGS14 rs7713145 a g 0.2807 -0.035 0.005 6.33E-12 3.13E-12 chr5 176781209 LMAN2 RGS14 rs4131290 a c 0.7042 0.0323 0.005 1.31E-11 8.72E-12 chr5 176780682 LMAN2 RGS14 rs2762918 t c 0.8388 0.0403 0.006 1.55E-11 5.20E-12 chr20 52744121 BCAS1 CYP24A1 rs2585448 c g 0.8387 0.0399 0.006 2.13E-11 7.24E-12 chr20 52744537 BCAS1 CYP24A1 rs2762919 t c 0.8387 0.0398 0.006 2.18E-11 7.48E-12 chr20 52744742 BCAS1 CYP24A1 rs2585449 a g 0.1614 -0.04 0.006 2.27E-11 7.77E-12 chr20 52744679 BCAS1 CYP24A1 rs2585447 t c 0.8388 0.0399 0.006 2.34E-11 7.95E-12 chr20 52744437 BCAS1 CYP24A1 rs2616278 t c 0.8387 0.0398 0.006 2.41E-11 8.24E-12 chr20 52744525 BCAS1 CYP24A1 rs2616279 t c 0.1623 -0.04 0.006 2.49E-11 8.52E-12 chr20 52743897 BCAS1 CYP24A1 rs532436 a g 0.2052 0.0359 0.005 2.83E-11 2.02E-07 chr9 136149830 ABO rs507666 a g 0.2052 0.0359 0.005 2.85E-11 4.20E-06 chr9 136149399 ABO rs33921462 a g 0.2675 -0.034 0.005 2.88E-11 3.99E-10 chr5 176814656 SLC34A1 rs635634 t c 0.2046 0.0357 0.005 3.40E-11 2.07E-07 chr9 136155000 ABO Y_RNA rs458385 a t 0.6881 0.0312 0.005 3.50E-11 2.50E-11 chr20 52715186 BCAS1 CYP24A1 rs458386 a t 0.6825 0.0312 0.005 3.78E-11 2.74E-11 chr20 52715187 BCAS1 CYP24A1 rs2616277 t c 0.1609 -0.039 0.006 4.07E-11 3.28E-12 chr20 52745040 BCAS1 CYP24A1 rs5030873 t c 0.6864 0.0324 0.005 1.09E-10 6.88E-11 chr5 176815124 SLC34A1 rs12659266 t c 0.2449 -0.032 0.005 1.10E-10 6.47E-11 chr5 176807197 SLC34A1 rs2519093 t c 0.202 0.0349 0.005 1.27E-10 7.49E-07 chr9 136141870 ABO

15 rs6556314 t c 0.3157 -0.032 0.005 1.33E-10 8.62E-11 chr5 176819252 SLC34A1 rs35610898 c g 0.3142 -0.032 0.005 1.36E-10 8.71E-11 chr5 176815766 SLC34A1 rs35630910 a c 0.7064 0.0297 0.005 1.46E-10 1.04E-10 chr5 176757841 MXD3 LMAN2 rs6862195 t g 0.3158 -0.032 0.005 1.70E-10 1.12E-10 chr5 176822512 SLC34A1 rs158529 t c 0.6591 0.0287 0.005 1.73E-10 9.01E-11 chr20 52721424 BCAS1 CYP24A1 rs6420095 a c 0.6842 0.0322 0.005 1.73E-10 1.15E-10 chr5 176821449 SLC34A1 rs6420094 a g 0.6843 0.032 0.005 1.76E-10 1.19E-10 chr5 176817636 SLC34A1 rs35716097 t c 0.2894 -0.032 0.005 1.77E-10 1.99E-06 chr5 176806636 SLC34A1 rs3812035 t g 0.3155 -0.032 0.005 1.79E-10 1.87E-07 chr5 176817143 SLC34A1 rs139415780 a g 0.1559 -0.038 0.006 1.81E-10 7.74E-11 chr20 52747748 BCAS1 CYP24A1 rs2762921 a g 0.156 -0.037 0.006 2.00E-10 8.82E-11 chr20 52749937 BCAS1 CYP24A1 rs158523 t c 0.6602 0.0286 0.005 2.04E-10 1.07E-10 chr20 52724309 BCAS1 CYP24A1 rs210068 a g 0.3392 -0.029 0.005 2.06E-10 1.12E-10 chr20 52726291 BCAS1 CYP24A1 rs158520 t c 0.3398 -0.029 0.005 2.12E-10 1.11E-10 chr20 52724718 BCAS1 CYP24A1 rs158527 c g 0.3399 -0.029 0.005 2.12E-10 1.11E-10 chr20 52722150 BCAS1 CYP24A1 rs10866705 a c 0.7438 0.0312 0.005 2.16E-10 1.24E-10 chr5 176801131 RGS14 SLC34A1 rs6556316 t c 0.3156 -0.032 0.005 2.19E-10 1.46E-10 chr5 176821375 SLC34A1 rs158526 a t 0.6601 0.0285 0.005 2.21E-10 1.16E-10 chr20 52722628 BCAS1 CYP24A1 rs158525 t c 0.6599 0.0285 0.005 2.24E-10 1.17E-10 chr20 52722810 BCAS1 CYP24A1 rs158528 a g 0.34 -0.029 0.005 2.25E-10 2.07E-11 chr20 52721956 BCAS1 CYP24A1 rs67190252 t c 0.2089 -0.038 0.006 2.54E-10 9.67E-11 chr5 176781905 LMAN2 RGS14 rs158521 t c 0.34 -0.028 0.005 2.74E-10 1.45E-10 chr20 52724665 BCAS1 CYP24A1 rs2762920 a g 0.1561 -0.037 0.006 2.86E-10 1.32E-10 chr20 52749823 BCAS1 CYP24A1 rs7447593 c g 0.6844 0.0321 0.005 3.87E-10 2.66E-10 chr5 176824137 SLC34A1 rs3812036 t c 0.2527 -0.03 0.005 1.08E-09 6.44E-10 chr5 176813404 SLC34A1 rs34604271 t c 0.2986 -0.028 0.005 1.11E-09 9.90E-10 chr5 176775851 LMAN2 rs62397245 c g 0.7688 0.0309 0.005 1.15E-09 8.47E-10 chr5 176750688 MXD3 LMAN2 rs6860069 t c 0.3092 -0.028 0.005 1.17E-09 6.53E-10 chr5 176755841 MXD3 LMAN2 rs34316179 a g 0.3184 -0.028 0.005 1.75E-09 8.39E-10 chr5 176779689 LMAN2 RGS14 rs55785724 a t 0.6815 0.03 0.005 1.82E-09 1.49E-09 chr5 176817583 SLC34A1

16 rs113955164 a g 0.2233 -0.032 0.005 1.97E-09 1.31E-09 chr5 176750988 MXD3 LMAN2 rs151014368 a g 0.2216 -0.031 0.005 2.01E-09 1.30E-09 chr5 176751059 MXD3 LMAN2 rs17479566 t c 0.2207 0.0313 0.005 2.01E-09 7.17E-07 chr9 71198014 RP11-274B18.4 rs4131289 a g 0.3211 -0.028 0.005 2.27E-09 1.15E-09 chr5 176780545 LMAN2 RGS14 rs2026351 a g 0.2221 0.0309 0.005 2.47E-09 1.00E-06 chr9 71194024 RP11-274B18.4 rs72716037 t c 0.2221 0.0308 0.005 2.58E-09 8.67E-07 chr9 71194967 RP11-274B18.4 rs954453 t c 0.2215 0.0306 0.005 3.37E-09 5.70E-09 chr9 71186301 RP11-274B18.4 rs630510 a g 0.4679 -0.025 0.004 3.40E-09 1.12E-05 chr9 136149581 ABO rs600038 t c 0.7738 -0.031 0.005 3.61E-09 6.41E-08 chr9 136151806 ABO Y_RNA rs630014 a g 0.468 -0.025 0.004 3.69E-09 8.83E-05 chr9 136149722 ABO rs12237628 t c 0.7763 -0.031 0.005 4.04E-09 8.26E-07 chr9 71187957 RP11-274B18.4 rs495828 t g 0.226 0.0307 0.005 4.18E-09 1.17E-07 chr9 136154867 ABO Y_RNA rs13362309 a g 0.6819 0.0277 0.005 4.31E-09 2.51E-09 chr5 176780137 LMAN2 RGS14 rs13190259 t c 0.3003 -0.028 0.005 4.38E-09 2.05E-09 chr5 176768110 LMAN2 rs651007 t c 0.226 0.0305 0.005 4.84E-09 6.48E-08 chr9 136153875 ABO Y_RNA rs1888722 t g 0.2128 0.0304 0.005 4.95E-09 1.02E-07 chr9 71218049 RP11-274B18.4 rs9925837 a g 0.8519 -0.035 0.006 5.11E-09 1.39E-04 chr16 79927303 RP11-345M22.3 RP11-148M9.1 rs579459 t c 0.7742 -0.031 0.005 5.23E-09 1.15E-06 chr9 136154168 ABO Y_RNA rs209961 t c 0.2847 -0.028 0.005 5.75E-09 2.72E-09 chr20 52715154 BCAS1 CYP24A1 rs649129 t c 0.2252 0.0305 0.005 5.78E-09 6.88E-08 chr9 136154304 ABO Y_RNA rs209960 t c 0.2852 -0.027 0.005 6.73E-09 4.80E-10 chr20 52716084 BCAS1 CYP24A1 rs13190105 a g 0.7105 0.0281 0.005 6.79E-09 3.41E-09 chr5 176768090 LMAN2 rs4976645 a g 0.321 -0.027 0.005 6.86E-09 3.61E-09 chr5 176777493 LMAN2 rs58256061 a g 0.2142 0.03 0.005 7.62E-09 1.76E-07 chr9 71199513 RP11-274B18.4 rs11644393 a t 0.1572 0.0335 0.006 7.70E-09 2.63E-04 chr16 79928128 RP11-345M22.3 RP11-148M9.1 rs2585412 a t 0.1494 -0.035 0.006 7.89E-09 4.04E-09 chr20 52758649 BCAS1 CYP24A1 rs4397143 a c 0.6789 0.0266 0.005 8.08E-09 4.31E-09 chr5 176777022 LMAN2 rs35582636 a c 0.2521 -0.029 0.005 8.18E-09 5.41E-09 chr5 176759778 LMAN2 rs111566909 a g 0.6918 0.0265 0.005 8.68E-09 5.35E-09 chr5 176767817 LMAN2 rs17479209 a g 0.2036 0.0302 0.005 8.84E-09 3.14E-06 chr9 71178959 RP11-274B18.4

17 rs10869048 a c 0.7959 -0.03 0.005 9.35E-09 4.40E-08 chr9 71178178 RP11-274B18.4 rs2762928 a t 0.8504 0.0343 0.006 1.05E-08 5.64E-09 chr20 52766085 BCAS1 CYP24A1 rs869950 a t 0.7968 -0.03 0.005 1.10E-08 3.95E-06 chr9 71179545 RP11-274B18.4 rs13157098 a g 0.144 -0.038 0.007 1.17E-08 3.98E-07 chr5 176802250 RGS14 SLC34A1 rs4627365 t c 0.1449 0.0342 0.006 1.21E-08 4.00E-05 chr16 79932213 RP11-345M22.3 RP11-148M9.1 rs11143037 t c 0.2158 0.0293 0.005 1.28E-08 3.26E-07 chr9 71208171 RP11-274B18.4 rs11649114 a g 0.8489 -0.033 0.006 1.58E-08 4.14E-04 chr16 79924857 RP11-345M22.3 RP11-148M9.1 rs28384289 t c 0.1507 0.033 0.006 1.61E-08 2.88E-04 chr16 79925400 RP11-345M22.3 RP11-148M9.1 rs11646139 a g 0.1506 0.033 0.006 1.62E-08 2.82E-04 chr16 79926321 RP11-345M22.3 RP11-148M9.1 rs9935686 a c 0.1506 0.033 0.006 1.62E-08 2.91E-04 chr16 79927218 RP11-345M22.3 RP11-148M9.1 rs11644400 t c 0.849 -0.033 0.006 1.65E-08 2.46E-05 chr16 79928186 RP11-345M22.3 RP11-148M9.1 rs9925120 c g 0.849 -0.033 0.006 1.72E-08 1.31E-07 chr16 79941214 RP11-345M22.3 RP11-148M9.1 rs4745134 t c 0.7836 -0.029 0.005 1.73E-08 6.08E-07 chr9 71205791 RP11-274B18.4 rs4745135 t c 0.7836 -0.029 0.005 1.74E-08 6.16E-07 chr9 71205800 RP11-274B18.4 rs11647905 c g 0.1508 0.033 0.006 1.75E-08 2.83E-04 chr16 79928065 RP11-345M22.3 RP11-148M9.1 rs9928026 a g 0.8493 -0.033 0.006 1.78E-08 2.87E-04 chr16 79927537 RP11-345M22.3 RP11-148M9.1 rs4078173 a g 0.8497 -0.033 0.006 1.83E-08 4.71E-06 chr16 79932332 RP11-345M22.3 RP11-148M9.1 rs2762932 t c 0.8501 0.0336 0.006 1.96E-08 2.06E-09 chr20 52768391 BCAS1 CYP24A1 rs13330604 a t 0.8497 -0.033 0.006 2.03E-08 2.93E-04 chr16 79926101 RP11-345M22.3 RP11-148M9.1 rs11641371 t c 0.1488 0.0331 0.006 2.03E-08 9.83E-09 chr16 79940725 RP11-345M22.3 RP11-148M9.1 rs11641245 c g 0.8482 -0.033 0.006 2.13E-08 4.71E-07 chr16 79940478 RP11-345M22.3 RP11-148M9.1 rs11641084 c g 0.1507 0.0327 0.006 2.16E-08 1.50E-07 chr16 79940800 RP11-345M22.3 RP11-148M9.1 rs9925132 t c 0.1508 0.0327 0.006 2.18E-08 1.58E-07 chr16 79941233 RP11-345M22.3 RP11-148M9.1 rs2616275 t c 0.215 -0.03 0.005 2.28E-08 1.15E-08 chr20 52745825 BCAS1 CYP24A1 rs7188156 t g 0.8491 -0.033 0.006 2.52E-08 2.49E-06 chr16 79938114 RP11-345M22.3 RP11-148M9.1 rs16951648 c g 0.1527 0.0323 0.006 2.67E-08 4.17E-06 chr16 79942149 RP11-345M22.3 RP11-148M9.1 rs12103233 a c 0.8487 -0.032 0.006 2.70E-08 4.67E-07 chr16 79941973 RP11-345M22.3 RP11-148M9.1 rs7189003 a t 0.1508 0.0324 0.006 2.79E-08 1.50E-07 chr16 79939801 RP11-345M22.3 RP11-148M9.1 rs28623477 a g 0.8482 -0.032 0.006 2.80E-08 2.17E-06 chr16 79942318 RP11-345M22.3 RP11-148M9.1 rs8053565 t g 0.1517 0.0323 0.006 2.83E-08 1.40E-07 chr16 79941892 RP11-345M22.3 RP11-148M9.1

18 rs8060495 a g 0.8493 -0.033 0.006 2.92E-08 4.13E-07 chr16 79942385 RP11-345M22.3 RP11-148M9.1 rs9313758 t g 0.678 0.025 0.005 2.98E-08 9.22E-09 chr5 176773091 LMAN2 rs35286975 c g 0.1519 0.0323 0.006 3.03E-08 2.20E-06 chr16 79938996 RP11-345M22.3 RP11-148M9.1 rs36102285 t c 0.3207 -0.025 0.005 3.03E-08 1.72E-08 chr5 176774018 LMAN2 rs4532376 a g 0.3208 -0.025 0.005 3.07E-08 8.42E-09 chr5 176774403 LMAN2 rs11142926 t g 0.2747 0.027 0.005 3.13E-08 2.24E-05 chr9 71176227 RP11-274B18.4 rs4745107 a g 0.7242 -0.027 0.005 3.29E-08 1.79E-08 chr9 71177110 RP11-274B18.4 rs11748912 a g 0.3224 -0.025 0.005 3.42E-08 1.89E-08 chr5 176765395 LMAN2 rs616154 t c 0.5209 0.0238 0.004 3.46E-08 3.28E-05 chr9 136150466 ABO rs4976643 a c 0.6774 0.0249 0.005 3.50E-08 1.96E-08 chr5 176760227 LMAN2 rs11949401 t c 0.6778 0.0248 0.005 3.53E-08 1.13E-08 chr5 176765989 LMAN2 rs6876677 t c 0.6777 0.0249 0.005 3.56E-08 2.02E-08 chr5 176755342 MXD3 LMAN2 rs12055081 t c 0.3222 -0.025 0.005 4.09E-08 2.32E-08 chr5 176757191 MXD3 LMAN2 rs4077451 a t 0.8391 -0.032 0.006 4.15E-08 7.96E-06 chr16 79931720 RP11-345M22.3 RP11-148M9.1 rs559723 a g 0.4871 -0.024 0.004 4.25E-08 2.55E-05 chr9 136150484 ABO rs6875461 t c 0.6783 0.0246 0.005 4.67E-08 2.70E-08 chr5 176754665 MXD3 LMAN2 rs11738681 a g 0.6774 0.0246 0.005 4.78E-08 2.72E-08 chr5 176761535 LMAN2 rs11749830 a g 0.323 -0.025 0.005 4.80E-08 2.63E-08 chr5 176766547 LMAN2 rs28581385 a t 0.852 -0.033 0.006 5.04E-08 2.81E-07 chr16 79942679 RP11-345M22.3 RP11-148M9.1 rs952436 t c 0.2082 0.0287 0.005 5.13E-08 4.53E-09 chr9 71219086 RP11-274B18.4 rs8058927 a g 0.151 0.032 0.006 5.83E-08 8.24E-10 chr16 79942562 RP11-345M22.3 RP11-148M9.1 rs2252316 a g 0.1396 -0.043 0.008 6.49E-08 5.41E-08 chr5 176837504 GRK6 rs28555498 a g 0.8453 -0.032 0.006 6.70E-08 8.13E-09 chr16 79942776 RP11-345M22.3 RP11-148M9.1 rs581107 t c 0.5602 0.0233 0.004 7.98E-08 1.25E-03 chr9 136147702 ABO rs475419 t c 0.5581 0.0232 0.004 8.68E-08 1.27E-03 chr9 136148231 ABO rs28446901 c g 0.8145 -0.03 0.006 9.71E-08 6.81E-04 chr9 136308796 ADAMTS13 rs645982 a g 0.4415 -0.023 0.004 9.93E-08 1.21E-03 chr9 136148409 ABO rs473533 t c 0.4415 -0.023 0.004 1.06E-07 1.28E-03 chr9 136148035 ABO rs476410 c g 0.5586 0.023 0.004 1.06E-07 1.31E-03 chr9 136148368 ABO rs659104 t g 0.4414 -0.023 0.004 1.10E-07 1.33E-03 chr9 136147823 ABO

19 rs660340 a g 0.4414 -0.023 0.004 1.11E-07 1.33E-03 chr9 136147553 ABO rs8054786 c g 0.8417 -0.031 0.006 1.17E-07 1.31E-07 chr16 79941881 RP11-345M22.3 RP11-148M9.1 rs62397242 a c 0.2379 -0.027 0.005 1.24E-07 8.71E-08 chr5 176738538 MXD3 rs500498 t c 0.4599 -0.023 0.004 1.25E-07 1.27E-03 chr9 136148647 ABO rs244730 a g 0.2435 -0.026 0.005 1.30E-07 8.18E-08 chr5 176539212 FGFR4 NSD1 rs11789139 c g 0.8258 -0.031 0.006 1.38E-07 2.44E-04 chr9 136185324 LCN1P2 SURF6 rs500499 c g 0.5378 0.0232 0.004 1.44E-07 1.57E-03 chr9 136148648 ABO rs12238108 t c 0.2083 0.0275 0.005 1.46E-07 6.86E-08 chr9 71221469 RP11-274B18.4 rs72813185 a c 0.2363 -0.027 0.005 1.48E-07 1.07E-07 chr5 176739110 MXD3 rs2252317 t g 0.8582 0.0414 0.008 1.54E-07 1.37E-07 chr5 176837490 GRK6 rs633862 t c 0.5512 0.0227 0.004 1.54E-07 1.49E-03 chr9 136155444 ABO Y_RNA rs244731 a g 0.2435 -0.026 0.005 1.63E-07 1.00E-07 chr5 176539679 FGFR4 NSD1 rs6123359 a g 0.9037 -0.041 0.008 1.79E-07 3.96E-08 chr20 52714706 BCAS1 CYP24A1 rs10869099 t c 0.7915 -0.027 0.005 1.83E-07 8.67E-08 chr9 71220831 RP11-274B18.4 rs244728 t c 0.2347 -0.027 0.005 2.17E-07 1.36E-07 chr5 176537407 FGFR4 NSD1 rs4637914 a t 0.2091 0.0269 0.005 2.49E-07 1.20E-07 chr9 71222223 RP11-274B18.4 rs244729 c g 0.755 0.0254 0.005 2.66E-07 1.72E-07 chr5 176539018 FGFR4 NSD1 rs67008484 t c 0.7613 0.0255 0.005 3.61E-07 2.60E-07 chr5 176735383 MXD3 rs7205649 t c 0.1428 0.0303 0.006 3.80E-07 1.23E-04 chr16 79939504 RP11-345M22.3 RP11-148M9.1 rs3105419 c g 0.2403 -0.025 0.005 3.98E-07 2.56E-07 chr5 176551396 FGFR4 NSD1 rs28579821 t c 0.2808 0.0325 0.006 4.22E-07 3.66E-07 chr7 155287014 CNPY1 rs8176690 a g 0.5862 0.022 0.004 4.48E-07 9.59E-05 chr9 136138317 ABO rs75719543 t c 0.9688 0.0809 0.016 4.64E-07 3.85E-07 chr12 85953371 RP11-408B11.2 RP11-629O19.1 rs244727 t c 0.2378 -0.026 0.005 4.71E-07 2.98E-07 chr5 176536549 FGFR4 NSD1 rs12235842 t c 0.2008 0.0271 0.005 4.86E-07 2.40E-07 chr9 71232678 RP11-274B18.4 rs7724098 a g 0.7594 0.0248 0.005 4.87E-07 3.77E-07 chr5 176659570 NSD1 rs9313752 a c 0.7595 0.0248 0.005 4.95E-07 3.77E-07 chr5 176645435 NSD1 rs13331920 c g 0.8582 -0.03 0.006 5.02E-07 4.21E-05 chr16 79935590 RP11-345M22.3 RP11-148M9.1 rs7193332 a g 0.8595 -0.03 0.006 5.19E-07 3.75E-05 chr16 79936215 RP11-345M22.3 RP11-148M9.1 rs2073828 a g 0.4132 -0.022 0.004 5.26E-07 6.44E-05 chr9 136137140 ABO

20 rs8176681 t c 0.5859 0.0218 0.004 5.44E-07 6.69E-05 chr9 136139754 ABO rs2073827 c g 0.4132 -0.022 0.004 5.45E-07 5.86E-05 chr9 136137133 ABO rs918459 a g 0.7594 0.0247 0.005 5.50E-07 4.34E-07 chr5 176669030 NSD1 rs11244061 t c 0.1172 0.0322 0.006 5.68E-07 3.52E-04 chr9 136153981 ABO Y_RNA rs35929305 t c 0.2187 0.0293 0.006 6.03E-07 6.22E-07 chr1 35092426 RP4-657M3.2 MIR552 rs2073826 t g 0.413 -0.022 0.004 6.17E-07 5.68E-05 chr9 136136963 ABO rs244707 a t 0.7623 0.025 0.005 6.27E-07 4.41E-07 chr5 176555636 FGFR4 NSD1 rs887258 c g 0.2195 0.0257 0.005 6.43E-07 2.46E-05 chr17 59479580 TBX2 rs1476781 t c 0.2203 0.0257 0.005 6.53E-07 7.81E-05 chr17 59476415 RP11-332H18.5 rs7721671 t c 0.24 -0.025 0.005 6.54E-07 4.96E-07 chr5 176630984 NSD1 rs2109019 a c 0.2201 0.0256 0.005 6.75E-07 1.61E-07 chr17 59475888 RP11-332H18.5 rs6561600 a g 0.7392 -0.024 0.005 6.90E-07 1.58E-06 chr13 33431045 PDS5B LINC00423 rs244726 a g 0.2507 -0.025 0.005 7.04E-07 4.56E-07 chr5 176536408 FGFR4 NSD1 rs11648875 t c 0.1415 0.0296 0.006 7.13E-07 2.45E-05 chr16 79939049 RP11-345M22.3 RP11-148M9.1 rs4976685 a g 0.6681 0.0221 0.005 7.21E-07 5.56E-07 chr5 176766177 LMAN2 rs6872952 a g 0.2561 -0.025 0.005 7.35E-07 5.34E-07 chr5 176650489 NSD1 rs12652509 a g 0.24 -0.025 0.005 7.50E-07 5.70E-07 chr5 176626441 NSD1 rs10037110 a t 0.2431 -0.024 0.005 7.67E-07 5.72E-07 chr5 176658066 NSD1 rs78463612 c g 0.9261 0.0422 0.009 8.09E-07 7.46E-07 chr6 100284585 Y_RNA MCHR2 rs4976677 a g 0.7599 0.0244 0.005 8.09E-07 6.14E-07 chr5 176629546 NSD1 rs4325578 a g 0.8561 -0.029 0.006 9.06E-07 3.11E-06 chr16 79936589 RP11-345M22.3 RP11-148M9.1 rs60715095 c g 0.8574 -0.029 0.006 9.07E-07 5.73E-05 chr16 79935768 RP11-345M22.3 RP11-148M9.1 rs28580467 c g 0.8577 -0.029 0.006 9.26E-07 4.11E-05 chr16 79934279 RP11-345M22.3 RP11-148M9.1 rs59520813 a g 0.143 0.0292 0.006 9.45E-07 9.70E-05 chr16 79936068 RP11-345M22.3 RP11-148M9.1 rs60280876 c g 0.1427 0.0292 0.006 9.61E-07 6.39E-05 chr16 79935778 RP11-345M22.3 RP11-148M9.1 rs647800 a g 0.4203 -0.021 0.004 9.64E-07 2.72E-03 chr9 136148000 ABO rs6556307 a g 0.2404 -0.024 0.005 9.68E-07 7.49E-07 chr5 176641434 NSD1 rs6872021 a g 0.7575 0.0242 0.005 1.05E-06 6.24E-07 chr5 176610012 NSD1 rs11740250 c g 0.2399 -0.024 0.005 1.07E-06 8.28E-07 chr5 176721272 NSD1 rs78097283 a t 0.9261 0.0414 0.009 1.11E-06 1.02E-06 chr6 100280629 Y_RNA MCHR2

21 rs79480825 a t 0.9261 0.0414 0.009 1.11E-06 1.02E-06 chr6 100280611 Y_RNA MCHR2 rs9907379 t c 0.2205 0.025 0.005 1.12E-06 8.99E-08 chr17 59489893 C17orf82 rs12479682 t c 0.1118 0.0329 0.007 1.13E-06 5.00E-07 chr20 52738937 BCAS1 CYP24A1 rs2762937 a g 0.8091 0.028 0.006 1.15E-06 2.24E-07 chr20 52778025 CYP24A1 rs1570669 a g 0.6581 -0.022 0.004 1.15E-06 4.74E-05 chr20 52774427 CYP24A1 rs7197544 t g 0.1433 0.0289 0.006 1.16E-06 1.41E-04 chr16 79937201 RP11-345M22.3 RP11-148M9.1 rs12480880 t c 0.8886 -0.033 0.007 1.19E-06 5.41E-07 chr20 52738743 BCAS1 CYP24A1 rs7203258 a g 0.143 0.0289 0.006 1.25E-06 1.18E-06 chr16 79937727 RP11-345M22.3 RP11-148M9.1 rs7204126 t c 0.143 0.0288 0.006 1.29E-06 5.55E-05 chr16 79937504 RP11-345M22.3 RP11-148M9.1 rs74038093 a t 0.1435 0.0293 0.006 1.30E-06 1.79E-03 chr16 79921081 RP11-345M22.3 RP11-148M9.1 rs2762934 a g 0.1831 -0.027 0.006 1.33E-06 5.55E-08 chr20 52771261 CYP24A1 rs115325566 t c 0.0788 -0.047 0.01 1.67E-06 2.85E-04 chr5 176807191 SLC34A1 rs13295207 a c 0.3538 0.0221 0.005 1.96E-06 4.78E-04 chr9 136075679 GBGT1 OBP2B rs887461 a t 0.7244 -0.023 0.005 2.19E-06 1.45E-06 chr7 90493764 CDK14 rs41340944 a g 0.0861 0.0369 0.008 2.22E-06 8.74E-07 chr10 70002020 RP11-153K11.3 rs751088 a t 0.1935 -0.027 0.006 2.25E-06 3.79E-07 chr20 52776892 CYP24A1 rs10869158 a g 0.7554 -0.024 0.005 2.37E-06 1.15E-04 chr9 71240051 RP11-274B18.4 rs11795370 t g 0.2777 0.0249 0.005 2.38E-06 5.83E-07 chr9 136059193 GBGT1 OBP2B rs8176668 a t 0.6072 0.0208 0.004 2.38E-06 1.85E-05 chr9 136144059 ABO rs58323452 a g 0.8196 -0.026 0.006 2.40E-06 1.26E-03 chr16 79928440 RP11-345M22.3 RP11-148M9.1 rs10823164 t c 0.9139 -0.037 0.008 2.44E-06 9.94E-07 chr10 70002705 RP11-153K11.3 KRT19P4 rs8068318 t c 0.7221 -0.022 0.005 2.51E-06 2.08E-04 chr17 59483766 TBX2 rs576123 t c 0.6536 -0.054 0.012 2.52E-06 2.17E-06 chr9 136144308 ABO rs933992 t c 0.1535 0.0268 0.006 2.58E-06 2.56E-06 chr3 69764203 RP11-444P10.1 MITF rs62071306 a c 0.344 0.0231 0.005 2.62E-06 5.98E-05 chr17 59476066 RP11-332H18.5 rs10998033 a g 0.914 -0.037 0.008 2.64E-06 1.04E-06 chr10 70002749 RP11-153K11.3 KRT19P4 rs7358139 t c 0.0861 0.0367 0.008 2.65E-06 1.08E-06 chr10 70002260 RP11-153K11.3 KRT19P4 rs7036642 a g 0.3931 -0.021 0.004 2.66E-06 1.01E-05 chr9 136144626 ABO rs9783267 c g 0.0861 0.0367 0.008 2.69E-06 1.07E-06 chr10 70002494 RP11-153K11.3 KRT19P4 rs73271107 a g 0.9138 -0.037 0.008 2.70E-06 1.08E-06 chr10 70002971 RP11-153K11.3 KRT19P4

22 rs12494375 c g 0.8465 -0.027 0.006 2.71E-06 2.69E-06 chr3 69765396 RP11-444P10.1 MITF rs7046674 t c 0.3928 -0.021 0.004 2.74E-06 2.29E-05 chr9 136147012 ABO rs8176686 t c 0.5869 0.0209 0.005 2.76E-06 NA chr9 136138776 ABO rs12952625 a g 0.7213 -0.022 0.005 2.91E-06 1.52E-05 chr17 59478644 TBX2 rs3802030 t g 0.2798 0.0223 0.005 2.93E-06 1.96E-06 chr7 90484578 CDK14 rs4855446 a g 0.1537 0.0266 0.006 3.13E-06 5.68E-06 chr3 69768996 RP11-444P10.1 MITF rs8074151 a g 0.7238 -0.022 0.005 3.15E-06 1.26E-03 chr17 59485017 TBX2 rs8078036 a g 0.7238 -0.022 0.005 3.18E-06 1.25E-03 chr17 59485120 TBX2 rs10869160 a g 0.2428 0.0232 0.005 3.18E-06 2.18E-05 chr9 71241343 RP11-274B18.4 rs10781095 a g 0.7571 -0.023 0.005 3.23E-06 1.85E-04 chr9 71240969 RP11-274B18.4 rs1057987 t c 0.7237 -0.022 0.005 3.28E-06 1.19E-03 chr17 59485555 TBX2 rs6767501 t c 0.1537 0.0266 0.006 3.33E-06 3.30E-06 chr3 69770411 RP11-444P10.1 MITF rs34446110 c g 0.7219 -0.022 0.005 3.35E-06 1.51E-03 chr17 59478354 TBX2 rs1000423 t c 0.7219 -0.022 0.005 3.36E-06 1.90E-03 chr17 59475642 RP11-332H18.5 rs4855322 a t 0.1536 0.0265 0.006 3.41E-06 3.38E-06 chr3 69769363 RP11-444P10.1 MITF rs2240736 t c 0.7254 -0.022 0.005 3.41E-06 1.56E-03 chr17 59485393 TBX2 rs2270114 c g 0.7095 -0.022 0.005 3.42E-06 1.12E-03 chr17 59478776 TBX2 rs11794634 t c 0.6613 -0.023 0.005 3.43E-06 1.22E-03 chr9 136090813 OBP2B LCN1P1 rs8176649 a g 0.3927 -0.02 0.004 3.45E-06 2.68E-05 chr9 136147295 ABO rs75973613 a t 0.9863 0.1088 0.023 3.46E-06 5.25E-07 chr13 46364453 SIAH3 rs7625281 t c 0.1536 0.0265 0.006 3.47E-06 3.45E-06 chr3 69768940 RP11-444P10.1 MITF rs4855443 a g 0.8463 -0.026 0.006 3.54E-06 3.56E-06 chr3 69766382 RP11-444P10.1 MITF rs4855445 t c 0.154 0.0264 0.006 3.55E-06 3.57E-06 chr3 69766596 RP11-444P10.1 MITF rs2289804 a g 0.9286 -0.04 0.009 3.55E-06 2.18E-06 chr10 69993032 RP11-153K11.3 rs8176682 t c 0.3924 -0.02 0.004 3.56E-06 1.88E-06 chr9 136139297 ABO rs8073698 t c 0.7238 -0.022 0.005 3.60E-06 1.29E-03 chr17 59484833 TBX2 rs4078174 a g 0.8245 -0.026 0.006 3.63E-06 5.66E-04 chr16 79932322 RP11-345M22.3 RP11-148M9.1 rs7873635 t c 0.3915 -0.021 0.004 3.74E-06 9.77E-06 chr9 136132012 ABO rs11697119 a g 0.3207 -0.022 0.005 3.83E-06 1.64E-06 chr20 52739292 BCAS1 CYP24A1 rs7215775 a g 0.2745 0.022 0.005 3.93E-06 1.67E-03 chr17 59484316 TBX2

rs12376362 a c 0.8057 -0.025 0.005 3.95E-06 8.75E-07 chr9 71172306 RP11-274B18.4 rs17597865 t c 0.0517 -0.051 0.011 3.98E-06 3.18E-06 chr15 39030873 C15orf53 RP11-326N17.1 rs8176691 t c 0.3919 -0.02 0.004 4.04E-06 1.28E-05 chr9 136138229 ABO rs8176702 a g 0.3915 -0.02 0.004 4.06E-06 1.30E-05 chr9 136136146 ABO rs4962040 a g 0.6081 0.0203 0.004 4.08E-06 1.76E-05 chr9 136133531 ABO rs10781092 a g 0.2915 0.0221 0.005 4.10E-06 1.55E-05 chr9 71239648 RP11-274B18.4 rs11791611 a g 0.3509 0.0215 0.005 4.15E-06 9.84E-04 chr9 136077441 GBGT1 OBP2B rs1411992 c g 0.8074 -0.025 0.005 4.17E-06 7.50E-07 chr9 71170702 RP11-274B18.4 rs4855444 t c 0.8407 -0.026 0.006 4.43E-06 4.56E-06 chr3 69766592 RP11-444P10.1 MITF rs1980513 t g 0.0769 0.0366 0.008 4.55E-06 5.26E-05 chr14 35581151 FAM177A1 rs9891115 a g 0.6476 -0.023 0.005 4.60E-06 2.41E-04 chr17 59482173 TBX2 rs13330024 a t 0.856 -0.027 0.006 4.68E-06 1.42E-03 chr16 79907901 RP11-345M22.3 RP11-148M9.1 rs7616567 a g 0.1539 0.0261 0.006 4.70E-06 4.72E-06 chr3 69766794 RP11-444P10.1 MITF rs62252967 t c 0.121 0.0285 0.006 4.83E-06 5.90E-06 chr3 69718556 RP11-444P10.1 MITF rs10869153 a c 0.7663 -0.023 0.005 4.87E-06 1.38E-04 chr9 71239400 RP11-274B18.4 rs729781 a g 0.276 0.0218 0.005 4.92E-06 1.52E-03 chr17 59486799 TBX2 rs558599 a c 0.4145 0.0238 0.005 4.94E-06 9.28E-04 chr9 136192551 LCN1P2 SURF6

P.value.FE: SNP P-value from fixed-effects model P.value.RE: SNP P-value from random-effects mode

Supplementary Table 3. Associations of top single nucleotide polymorphisms with ln- transformed intact FGF23 concentration (exclusion of CHS and OPRA GWAS from meta- analysis)

FGF23 Model 1 Model 2† FGF23 Nearest Other Increasing (N=13,670) (N=12,519) SNP Chr Position Increasing geneλ Allele Allele Allele β‡ (SE) p-value β (SE) p-value Frequencyα 0.80 0.057 0.054 rs17216707 CYP24A1 20 52732362 T C 1.0 x 10-25 1.5 x 10-23 (0.006) (0.005) 0.37 0.037 0.032 rs2769071 ABO 9 136145974 G A 7.1 x 10-16 1.3 x 10-4 (0.005) (0.008) 0.73 0.038 0.037 rs11741640 RGS14 5 176792743 G A 1.6 x 10-14 3.6 x 10-14 (0.005) (0.005) 0.22 0.032 0.044 rs17479566 LINC01506 9 71198013 T C 3.8 x 10-9 3.8 x 10-6 (0.005) (0.009) 0.13 0.032 0.032 rs9925837 LINC01229 16 79927303 G A 2.2 x 10-7 1.0 x 10-7 (0.006) (0.006)

Chr = chromosome Only top SNP from each region shown. Model 1 includes age, sex and first 10 principal components of ancestry Model 2 additionally adjusts for BMI, eGFR and eGFR-squared. †Results for Model 2 were not available for the Indiana Sisters Study αMinor allele frequency data from 1000 Genome Phase 1 genotype data ‡Beta-estimates are interpreted as the relative difference in FGF23 concentration per minor allele, e.g. 0.055 is a 5.5% higher FGF23 concentration per additional allele

Supplementary Table 4. Genetically predicted gene expression results from GWAS summary statistics

Tissue Gene Chr Z -score P value Variance R2 N SNPs Model N Skeletal Muscle RGS14* 5 7.956 1.78E-15 0.104 0.166 26 36 Subcutaneous Adipose RGS14* 5 7.017 2.26E-12 0.079 0.178 4 4 Heart: Atrial Appendage RGS14* 5 7.005 2.47E-12 0.18 0.305 14 14 Aortic Artery RGS14* 5 6.887 5.70E-12 0.279 0.378 30 31 Transformed Fibroblast Cells RGS14* 5 6.748 1.50E-11 0.418 0.568 22 23 Heart: Left Ventricle RGS14* 5 6.219 5.00E-10 0.192 0.313 22 23 Coronary Artery RGS14* 5 4.649 3.34E-06 0.149 0.21 27 29 Subcutaneous Adipose TBX2* 17 4.581 4.62E-06 0.046 0.024 32 34 Subcutaneous Adipose HDDC2 6 4.527 5.97E-06 0.135 0.125 46 50 Skeletal Muscle NSD1* 5 -4.33 1.50E-05 0.012 0.035 6 6 Skeletal Muscle TBX2* 17 4.043 5.27E-05 0.008 0.031 6 6 Skeletal Muscle EIF3F 11 -4.04 5.28E-05 0.078 0.145 23 24 Subcutaneous Adipose ARHGAP29 1 3.981 6.86E-05 0.048 0.05 40 42 Subcutaneous Adipose EIF3F 11 -3.96 7.40E-05 0.084 0.164 17 18 Subcutaneous Adipose JPH2 20 -3.93 8.36E-05 0.036 0.023 33 38 Aortic Artery SNX7 1 -3.91 9.16E-05 0.082 0.168 14 14 Skeletal Muscle MXD3* 5 3.908 9.30E-05 0.065 0.095 40 41 Transformed Fibroblast Cells KCTD6 3 -3.85 1.19E-04 0.027 0.01 20 20 Heart: Atrial Appendage MXD3* 5 3.821 1.33E-04 0.062 0.148 15 15 Skeletal Muscle RPL29 3 3.809 1.39E-04 0.232 0.322 33 47 Skeletal Muscle ZNF404 19 3.789 1.51E-04 0.062 0.163 34 38 Skeletal Muscle WARS2 1 3.756 1.73E-04 0.266 0.41 21 21 Whole Blood Cells EIF3F 11 -3.72 1.98E-04 0.093 0.149 25 25 Transformed Fibroblast Cells LTBP4 19 3.714 2.04E-04 0.081 0.091 39 40 Transformed Fibroblast Cells DOK5* 20 -3.7 2.17E-04 0.002 0.007 9 9 Spleen REXO1 19 -3.66 2.55E-04 0.078 0.049 26 26 Subcutaneous Adipose WARS2 1 3.631 2.82E-04 0.347 0.497 18 19 Skeletal Muscle PDLIM7* 5 3.63 2.83E-04 0.202 0.268 28 35 Heart: Atrial Appendage KDM4A 1 3.626 2.87E-04 0.002 0.052 4 4 Liver SMYD2 1 3.616 2.99E-04 0.1 0.139 25 26 Heart: Atrial Appendage PVRL2 19 3.6 3.18E-04 0.146 0.126 47 48 Transformed Fibroblast Cells ENO1 1 -3.57 3.53E-04 0.008 0.02 14 15 Transformed Fibroblast Cells PPP2CB 8 -3.56 3.65E-04 0.052 0.075 30 33 Transformed Fibroblast Cells FAM124A 13 3.533 4.11E-04 0.01 0.016 5 6 Subcutaneous Adipose PYGL 14 -3.53 4.12E-04 0.029 0.039 29 32 Skeletal Muscle SVIP 11 -3.53 4.12E-04 0.162 0.191 33 41 Subcutaneous Adipose BTBD19 1 3.532 4.13E-04 0.005 0.028 10 13 Skeletal Muscle CCNDBP1 15 3.496 4.73E-04 0.007 0.007 7 7 *represents genes which localize to top GWAS signals (Table 2) Performance prediction R2 Variance of the gene’s predicted expression, calculated as W’x G x W (where W is the vector of SNP weights in a gene’s model, W’ is its transpose, and G is the covariance matrix) Number of SNPs included in the prediction model for that gene available in the FGF23GWAS summary statistics Number of SNPs used to construct the prediction model for the gene in the tissue of interest using the GTExdata

SIGNIFICANCE STATEMENT

Fibroblast growth factor 23 (FGF23), a bone-derived hormone that regulates phosphorus and vitamin D metabolism, contributes to the pathogenesis of mineral and bone disorders in CKD and is an emerging cardiovascular risk factor. The authors performed a meta-analysis of genome- wide association studies of circulating FGF23 concentrations among 16,624 individuals of Euro- pean ancestry from seven cohort studies. After adjusting for age, sex, study site, and principal components of ancestry, they found that common genetic variants are associated with differences in circulating FGF23 concentration; several are closely linked with enzymes, transporters, and receptors known to be critical to vitamin D metabolism and regulation of phosphate levels. Future study of such variants may help illuminate the mechanism and clinical implications of FGF23’s role in vitamin D and phosphate homeostasis.